Understanding the effect of soluble fibres on the

Understanding the effect of soluble fibres on the hydrolysis of starch

and the diffusion of glucose during simulated human digestion

by

Hrvoje Fabek

A Thesis

Presented to

The University of Guelph

In partial fulfilment of requirements

for the degree of

Doctor of Philosophy

in

Food Science

Guelph, Ontario, Canada

© Hrvoje Fabek, April, 2015

ABSTRACT

UNDERSTANDING THE EFFECT OF SOLUBLE FIBRES ON THE

HYDROLYSIS OF STARCH AND THE DIFFUSION OF GLUCOSE DURING

SIMULATED HUMAN DIGESTION

Hrvoje Fabek Advisor:

University of Guelph, 2015 Professor H.D. Goff

Dietary fibre consumption is associated with a wide range of health benefits

including reductions in postprandial glycemia. It is widely accepted among

researchers that the effect is due to an increase in viscosity that fibres exert along the

gastrointestinal tract; however, the mechanism remains to be elucidated. In this study,

protein-starch solutions were fortified with four types of soluble fibre – xanthan gum

(XG), guar gum (GG), soluble flaxseed gum (SFG) and soy soluble polysaccharide

(SSPS). Subsequently, all solutions were passed through a three-stage in vitro

digestion model. Rheological investigations showed that all solutions behaved as

pseudoplastic shear-thinning materials and exhibited solid-like behaviour.

Quantitative analyses of reducing sugar release (RSR) demonstrated an inverse trend

between RSR and digesta viscosity. Glucose release measurements revealed that even

low viscosity systems were effective at lowering the hydrolysis of starch. Analysis of

glucose diffusion using a dialysis system indicated a lesser effect and only the fibres

with a measurable G’ and higher viscous component inside the simulated small

intestinal stage were able to attenuate glucose diffusion. Native (uncooked) tapioca

starch granules were separated, purified and dried at different times of digestion and

analyzed for changes in particle size using light scattering and starch morphology

using scanning electron microscopy (SEM) and light microscopy. Most granules were

in the range of 5-50µm, with an apparent reduction in particle size as hydrolysis

continued. SEM images of starch granules revealed smooth surfaces in solution and

substantial degradation (exo- and endo-corrosion, with some granules becoming

completely hydrolysed) as the treatments progressed through the in vitro digestion.

The progression of morphological changes was attenuated in granules extracted from

the digesta of XG and GG. Light scattering and microscopy also revealed an

aggregation effect between granules extracted from the XG and GG treatments. The

results of the study indicated that the mechanism by which soluble fibres are able to

moderate the in vitro glycemic response may be multi-faceted. Moreover, they may

not be due to viscosity alone as other effects may be important, including the ability of

fibres to create and maintain gel networks inside the lumen and allow starch granule

aggregation.

iv

Acknowledgements

I would like to express my appreciation to my advisor Dr. Douglas Goff,

without whom this project would not have been possible. The experience, knowledge,

and skills that I have gained through the completion of this project are invaluable.

Thank you for the guidance and incessant support you have offered throughout my

graduate degrees.

Special thanks to my two advisory committee members, Dr. Alison Duncan

and Dr. Amanda Wright. Your knowledge and contribution to this project are greatly

appreciated. Thank you for offering me your time, feedback and expertise throughout

this project.

Thank you to Dr. Sandy Smith who helped me with all the microscopy

analyses and providing training on the SEM, CLSM and light mic.

Thanks to all my labmates and friends who have helped make the time spent

on this project enjoyable and a special thanks to Dr. Fernanda Peyronel for her efforts

in maintaining the laboratory where I conducted most of my research. Also, thank you

to Ms. Tricia Anderson, Ms. Anne Ingram and Ms. Leona Varga-Lowes for their help

on the administrative side of the project.

Finally, I would like to thank and dedicate this thesis to my family who have

been supportive of me every step of the way. To my beautiful wife Danielle and our

two beautiful children - our son Luka and daughter Arwyn (born 4 days before my

defence) – thank you for making every day the happiest day of my life, I love you

guys so much!! Thank you to my parents for the never-ending support offered to me

throughout my life - hvala na svemu!

v

Table of Contents

1. Introduction ........................................................................................................................ 1

1.1. Importance of dietary fibre and research focus .......................................................... 1

1.2. Overall Objectives ..................................................................................................... 5

2. Literature Review ............................................................................................................... 7

2.1. Digestion of foods ...................................................................................................... 7

2.1.1. Gastrointestinal physiology...................................................................................... 7

2.1.2. Carbohydrate metabolism ...................................................................................... 12

2.1.3. Glucose absorption ................................................................................................. 14

2.1.4. Review of in vitro digestion ................................................................................... 16

2.1.5. In vitro digestion methods related to starch digestion ........................................... 23

2.2. Dietary Fibre ............................................................................................................ 29

2.2.1. Dietary fibre definition .......................................................................................... 29

2.2.2. Health benefits of dietary fibre .............................................................................. 31

2.3. Soluble fibres employed in the study ....................................................................... 42

2.3.1. Guar Gum............................................................................................................... 42

2.3.2. Xanthan Gum ......................................................................................................... 46

2.3.3. Flaxseed Gum ........................................................................................................ 50

2.3.4. Soy soluble polysaccharide .................................................................................... 52

2.3.5. Summary Comment ............................................................................................... 55

3. The effect of in vitro digestive processes on the viscosity of dietary fibres and their

influence on glucose diffusion ................................................................................................. 56

3.1. Introduction .............................................................................................................. 57

3.2. Materials and methods ............................................................................................. 60

3.2.1. Materials ................................................................................................................ 60

3.2.2. Soluble flaxseed gum extraction ............................................................................ 61

3.2.3. Food matrix preparation ........................................................................................ 61

3.2.4. Viscosity measurements ......................................................................................... 62

3.2.5. In vitro digestion .................................................................................................... 63

3.2.6. Glucose release measurements .............................................................................. 63

3.2.7. Statistical Analysis ................................................................................................. 64

3.3. Results and discussion ............................................................................................. 64

3.3.1. Effect of in vitro digestion on solution viscosity .................................................... 64

vi

3.3.2. Glucose diffusion during in vitro digestion ............................................................ 71

3.4. Conclusion ............................................................................................................... 75

4. Effect of soluble fibre inclusion on starch hydrolysis and glucose mobility during

simulated small intestinal digestion ......................................................................................... 77

4.1. Introduction .............................................................................................................. 78

4.2. Materials and methods ............................................................................................. 81

4.2.1. Materials ................................................................................................................ 81

4.2.2. Food matrix preparation ........................................................................................ 82

4.2.3. Dynamic viscoelasticity and viscosity measurements ............................................ 83

4.2.4. In vitro digestion .................................................................................................... 83

4.2.5. Reducing sugar determination ............................................................................... 85

4.2.6. Glucose release determination inside simulated digesta ....................................... 85

4.2.7. Glucose diffusion inside a dialysis system ............................................................. 86

4.2.8. Statistical Analysis ................................................................................................. 86

4.3. Results and discussion ............................................................................................. 87

4.3.1. Study treatment rheology during in vitro digestion ............................................... 87

4.3.2. Reducing sugar release during simulated small intestinal digestion ...................... 99

4.3.3. Effects of viscous fibres on glucose release during simulated small intestinal

digestion ......................................................................................................................... 102

4.4. Conclusion ............................................................................................................. 108

5. Simulated intestinal hydrolysis of native tapioca starch: understanding the effect of

soluble fibre ........................................................................................................................... 110

Hrvoje Fabeka and H. Douglas Goff

a, .................................................................................... 110

5.1. Introduction ............................................................................................................ 111

5.2. Materials and methods ........................................................................................... 115

5.2.1. Materials .............................................................................................................. 115

5.2.2. Sample preparation .............................................................................................. 116

5.2.3. In vitro digestion .................................................................................................. 116

5.2.4. Viscosity measurements ....................................................................................... 117

5.2.5. Isolation and purification of starch granules for morphological study ............... 117

5.2.6. Particle Size Analysis ........................................................................................... 118

5.2.7. Scanning electron microscopy (SEM) .................................................................. 119

5.2.8. Light microscopy .................................................................................................. 119

5.2.9. Statistical analysis ............................................................................................... 119

vii

5.3. Results and discussion ........................................................................................... 120

5.3.1. Flow behaviour .................................................................................................... 120

5.3.2. Particle size distribution of hydrolysed starch granules ....................................... 122

5.3.3. Starch Morphology .............................................................................................. 130

5.3.3.1. Scanning electron microscopy (SEM) ............................................................... 130

5.3.3.2. Light microscopy of digested starch granules .................................................. 141

5.4. Conclusion ............................................................................................................. 146

6. Study strengths and limitations ...................................................................................... 148

7. Conclusion and future directions ................................................................................... 150

References .............................................................................................................................. 155

viii

List of Tables

Table 3-1. Study treatment formulation (% w/w) for control solution (CS), xanthan

gum (XG)-, guar gum (GG)-, soluble flaxseed gum (SFG)-, and soluble soy

polysaccharide (SSPS)-fortified solutions.. .............................................................. 62

Table 3-2. Consistency index (K) and flow behaviour index (n) from the Power Law model of

guar gum-, xanthan gum-, flaxseed gum-, and soluble soy polysaccharide (SSPS)-

fortified solutions before and after in vitro digestion. .................................................... 70

Table 3-3. Apparent viscosity (Pa.s) at 50s-1 of guar gum-, xanthan gum-, flaxseed gum-, and

soluble soy polysaccharide (SSPS)-containing solutions after 4 h in vitro digestion. .... 72

Table 3-4. Normalized Sugar Diffusion (NSD) of solutions mixed with fibre compared with

the control ....................................................................................................................... 73

Table 4-1. Study treatment formulation (% w/w) for control solution (CS), xanthan gum

(XG)-, guar gum (GG)-, flaxseed gum (FG)-, and soluble soy polysaccharide

(SSPS)-fortified solutions ............................................................................................... 83

Table 4-2. Chemical composition of simulated salivary fluid1 ........................................................ 84

Table 4-3. Mean values for viscoelastic measurements storage (G') and loss (G") moduli and

tan δ measured at 1 Hz of xanthan gum-, guar gum-, flaxseed gum, and soluble soy

polysaccharide-fortified solutions ................................................................................... 93

Table 4-4. Consistency index (K, Pa.s) values of the Power Law model for control, xanthan

gum-, guar gum-, soy soluble polysaccharide (SSPS)-, and flaxseed gum-fortified

treatments in solution, following simulated salivary (SP), gastric (GP) and small

intestinal (SIP) phases of digestion. ................................................................................ 97

Table 4-5. Behaviour index (n) values of the Power Law model for control, xanthan gum-,

guar gum-, soy soluble polysaccharide (SSPS)-, and flaxseed gum-fortified


intestinal phases of digestion. ......................................................................................... 98

Table 4-7. Calculated slopes of the lines of regression (SLREG) of reducing sugar release for

control, xanthan gum-, guar gum-, flaxseed gum, and soy soluble polyssacharide

(SSPS)-fortified treatments. .......................................................................................... 100

Table 4-8. Percentage of starch hydrolysed (as measured by the amount of glucose released)

after in vitro digestion of control (no gum), xanthan gum-, guar gum, flaxseed

gum, and soy soluble polysaccharide (SSPS)-fortified solutions. ................................ 103

ix

Table 4-9. Final glucose concentrations measured inside the dialysate and calculated slopes of

the lines of regression (SLREG) of glucose release for control, xanthan gum-, guar

gum-, flaxseed gum, and soy soluble polysaccharide (SSPS)-fortified treatments

during 5 h in vitro digestion. ......................................................................................... 107

Table 5-1. Power Law parameters of control (no fibre) xanthan gum (XG)-, guar gum (GG)-,

soluble flaxseed gum (SFG)-, and soy soluble polysaccharide (SSPS)-fortified

treatments of the digesta isolated at the beginning (T0), after 1 h (T1), 2 h (T2),

and at the end (T3) of simulated small intestinal digestion. ......................................... 121

Table 5-2. Particle size as defined by the volume weighted mean d(4,3) of granules isolated

from in vitro digesta before the addition of amyloglucosidase (T0) after 1 h (T1), 2

h (T3), and 4 h (T3) of simulated small intestinal digestion. ........................................ 123

x

List of Figures

Figure 2-1. A comparison of solid and liquid components remaining in the stomach after

ingestion of a meal (Adapted from Camilleri, Malagelada, and Brown, 1985). .......... 10

Figure 2-2. GLUT2 model showing intestinal aborption of glucose following a meal (Adapted

from Kellett and Brot-Laroche, 2005). ......................................................................... 15

Figure 2-3. Schematic diagram of the dynamic gastric model. Numbers illustrate passage of

meal contents from the (1) the main body to the (2) the antrum to allow for

processing using (3) the piston and barrel to allow for (4) analysis (Adapted from

Vardakou et al., 2011). ................................................................................................. 20

Figure 2-4. Proposed flow diagram for a standardised in vitro digestion method (Adapted

from Minekus et al., 2014) ........................................................................................... 22

Figure 2-5. Morphology of (a) raw maize starch granules isolated from in vivo digesta and (b)

starch granules isolated from in vitro digesta (Adapted from Hasjim et al., 2010). ..... 26

Figure 2-6. Diagram of the potential effect of viscous soluble dietary fibre (SDF) on reducing

micelle absorption (Adapted from Gunness and Gidley, 2010). .................................. 36

Figure 2-7. Flow diagram illustrating guar gum production (Adapted from Wielinga, 2010). ....... 43

Figure 2-8. Chemical structure of the galactomannan backbone (Adapted from Wielinga,

2010). ............................................................................................................................ 44

Figure 2-9. Flow diagram illustrating xanthan gum production (Adapted from Garcia-Ochoa

et al., 2000). .................................................................................................................. 47

Figure 2-10. Structure of xanthan gum (Adapted from Garcia-Ochoa et al., 2000). ....................... 48

Figure 2-11. Structure of SSPS (adapted from Nakamura, et al., 2002). ......................................... 53

xi

Figure 3-1. Viscosity of guar gum-, xanthan gum-, soluble soy polysaccharide (SSPS)-, and

flaxseed gum- fortified solutions, as a function of shear rate (s-1

). .............................. 65



) following 1h

simulated gastric digestion. .......................................................................................... 66



), following a

two-stage in vitro digestion. ......................................................................................... 67

Figure 3-4. Viscosity of guar gum and guar gum-fortified solution containing protein and

starch, following simulated in vitro digestion (curves are overlapping). ..................... 68

Figure 3-5. Viscosity of xanthan gum and xanthan gum-fortified solution containing protein

and starch, following simulated in vitro digestion. ...................................................... 68

Figure 3-6. Viscosity of flaxseed gum and flaxseed gum-fortified solution containing protein

and starch, following simulated in vitro digestion. ...................................................... 69

Figure 3-7. Viscosity of soy soluble polysaccharide (SSPS) and SSPS-fortified solution

containing protein and starch, following simulated in vitro digestion. ........................ 69

Figure 3-8. Available glucose (µg/mL) in the dialysate during 180 min in vitro small

intestinal digestion of control (no fibre), soy soluble polysaccharide (SSPS)-,

flaxseed gum-, guar gum-, and xanthan gum-fortified solutions. ................................ 71


intestinal digestion of control (no fibre), soy soluble polysaccharide (SSPS)-,

flaxseed gum-, guar gum-, and xanthan gum- fortified solutions when starch was

replaced with D-glucose. ............................................................................................... 75

Figure 4-1. Storage (G') and loss moduli (G") of xanthan gum-fortified treatments – that is

(a) in solution, (b), following simulated gastric digestion, and (c) at the end of 3-

stage in vitro digestion – as a function of frequency with an oscillating stress of

0.5 Pa. ........................................................................................................................... 88

xii

Figure 4-2. Storage (G') and loss moduli (G") of guar gum-fortified treatments – that is (a) in

solution, (b), following simulated gastric digestion, and (c) at the end of 3-stage

in vitro digestion – as a function of frequency with an oscillating stress of 0.5 Pa. .... 89

Figure 4-3. Storage (G') and loss moduli (G") of flaxseed gum-fortified treatments – that is

(a) in solution, (b), following simulated gastric digestion, and (c) at the end of 3-

stage in vitro digestion – as a function of frequency with an oscillating stress of

0.5 Pa. ........................................................................................................................... 90

Figure 4-4. Storage (G') and loss moduli (G") of soluble soy polysaccharide-fortified

treatments – that is (a) in solution, (b), following simulated gastric digestion, and

(c) at the end of 3-stage in vitro digestion – as a function of frequency with an

oscillating stress of 0.5 Pa. ........................................................................................... 91

Figure 4-5. Flow behaviour of control (no gum), xanthan gum-, guar gum-, soy soluble

polysaacharide (SSPS)-, and flaxseed gum-fortified solutions across a shear rate

range of 10 to 200 s-1

. ................................................................................................... 95

Figure 4-6. Flow behaviour of control (no gum), xanthan gum-, guar gum-, soy soluble

polysaacharide (SSPS)-, and flaxseed gum-fortified solutions - that follow

simulated, (a) salivary, (b) gastric, and (c) small intestinal digestion, across a

shear rate range of 10 to 200 s-1

. ................................................................................... 96

Figure 4-7. Reducing sugars released, expressed as mg/g sample of control (no fibre), soy

soluble polysaccharide (SSPS)-, flaxseed gum-, guar gum-, and xanthan gum-

fortified solutions, during 5 h simulated intestinal digestion. ...................................... 99

Figure 4-8. Slopes of the lines of regression (SLERG) of reducing sugar release for control,

xanthan gum-, guar gum-, flaxseed gum, and soy soluble polyssacharide (SSPS)-

fortified treatments versus apparent viscosity (50s-1

) during simulated small

intestinal digestion. ..................................................................................................... 101

Figure 4-9. Percentage of starch hydrolysis during 5 h simulated small intestinal digestion of

control (no fibre), soy soluble polysaccharide (SSPS)-, flaxseed gum-, guar gum-

, and xanthan gum-fortified solutions (measured by the concentration of glucose). .. 102

Figure 4-10. In vitro diffusion of glucose during 5 h simulated small intestinal digestion of

control (no fibre), soy soluble polysaccharide (SSPS)-, flaxseed gum-, guar gum-

, and xanthan gum-fortified solutions. ........................................................................ 106

xiii

Figure 5-1. Flow behaviour of native starch/skim milk powder solutions fortified with 4%

xanthan gum, 3% guar gum, 7% soluble flaxseed gum, and 20% soy soluble

polysaccharide (SSPS). .............................................................................................. 120

Figure 5-2. Particle size distribution of starch granules isolated from the in vitro digesta of

control- and fibre-fortified solutions at the beginning of simulated small

intestinal hydrolysis. ................................................................................................... 124


control- and fibre-fortified solutions after 1 hr of simulated small intestinal

hydrolysis. .................................................................................................................. 125


control- and fibre-fortified solutions after 2 h of simulated small intestinal

hydrolysis. .................................................................................................................. 126


control- and fibre-fortified solutions at the end of simulated small intestinal

hydrolysis. .................................................................................................................. 127

Figure 5-6. Scanning electron microscopy of tapioca starch granules isolated from control

(C)-, soy soluble polysaccharide (SSPS)-, soluble flaxseed gum (SFG)-, guar

gum (GG)-, and xanthan gum (XG)-fortified solutions. ............................................ 131

Figure 5-7. Scanning electron micrographs of tapioca starch aggregates isolated from guar

gum (GG)- and (XG)-fortified solutions. ................................................................... 132

Figure 5-8. Morphology of granules isolated from the in vitro digesta of the control and fibre

foritifed treatments at the beginning of simulated small intestinal digestion. ............ 134

Figure 5-9. Morphology of granules isolated from the in vitro digesta of the control and fibre

fortified treatments after 1 h of simulated small intestinal digestion. ........................ 136

Figure 5-10. Morphology of granules isolated from the in vitro digesta of the control and

fibre fortified treatments after 2 h of simulated small intestinal digestion. ................ 137

xiv

Figure 5-11. Morphology of granules isolated from the in vitro digesta of the control and

fibre fortified treatments at the end of in vitro digestion. ........................................... 139

Figure 5-12. Light microscopy of starch isolated from the digesta of control (C), soy soluble

polysaccharide (SSPS), soluble flaxseed gum (SFG), guar gum (GG) and xanthan

gum (XG) treatments at the beginning (T1), after 1 h (T2), after 2 h (T3) and at

the end of simulated small intestinal digestion (T4)................................................... 141

xv

List of Acronyms and symbols

XG Xanthan gum

GG Guar gum

SFG Soluble flaxseed gum

SSPS Soy soluble polysaccharide

RSR Reducing sugar release

NSD Normalized sugar diffusion

GIT Gastrointestinal tract

DGM Dynamic gastric model

CS Control solution

G’ Storage modulus

G” Loss modulus

σ Shear stress

K Consistency index

n Flow behaviour index

SEM Scanning electron microscopy

1

1. Introduction

1.1. Importance of dietary fibre and research focus

Mankind has been enduring global change for centuries, whether they are

political, economic or societal; yet the single most invariable and important factor in

our everyday lives remains to be healthy. For decades the link between proper

dietary eating patterns and improvement in overall health and wellness has been

perpetuated and strengthened through research studies and government

recommendations. The Public Health Agency of Canada (2013) has developed what

is known as the “Preventing Chronic Disease Strategic Plan” in order to provide a

roadmap to healthy living. The front runners for this strategy include two

controllable factors, those being regular exercise and healthy eating habits. This link

between food and health is being perpetuated in the minds of consumers and also

legislators who are incessantly looking for long-term reduction in health care costs.

Despite this increase in awareness there is an interminable upsurge in

disease trends in North America. To exemplify this, self-reported obesity trends have

increased from 6.2% in 1985 to 16.9% in 2007. Furthermore, the number of people

diagnosed with type II diabetes (T2D) in Canada alone has increased from 3.3% in

1999 to 5.6% in 2009 and if these incidence rates continue on this path the expected

number of people diagnosed with T2D is expected to reach 3.7 million by 2019.

According to the World Health Organization, T2D affects more than 380 million

people worldwide and this number is expected to double by the year 2025 (World

Health Organization, 2003). Beyond the health implications that these trends have

they also bear a significant economic impact, as it has been shown that populations

2

with diabetes have nearly four times greater annual per capita health care costs than

populations without (Public Health Agency of Canada, 2011). Nonetheless,

Canadian consumers are consistently being advised on the importance of exercise

and eating a healthy balanced diet, with the primary focus being fat reduction and

consuming foods rich in dietary fibre. The latter component is one that has gained a

great deal of attention from not only consumers but also health practitioners,

researchers, legislators and food manufacturers, due to the multitude of health

benefits associated with the consumption of dietary fibre.

Recommended intake levels for dietary fibre are based on the level observed

to protect against coronary heart disease (Institute of Medicine, 2005). In addition to

this, consumption of fibre-rich foods is associated with other physiological effects

including increases in satiety, reductions in blood lipids, specifically triglycerides

and low-density lipoprotein cholesterol, decreases in blood glucose levels, as well as

enhanced gastrointestinal immunity and overall colonic health (Jenkins, Leeds,

Gassull, Cochet, & Alberti, 1977; Gunness & Gidley, 2010; Brownlee, 2011).

Globally, government agencies such as the Canadian Diabetes Association,

American Diabetes Association, British Diabetes Association, and European

Association for the Study of Diabetes all have made strong nutritional

recommendations to include dietary fibre as part of a regular diet for people

suffering from T2D (Anderson, Randles, Kendall, and Jenkins, 2004). Currently, the

recommended intake levels in North America are 38 g/d and 25 g/d for men and

women, respectively. However, recent surveys from nutrient intakes from foods

indicate that actual intake levels fall between 16.5 to 19.4 g/d for men and 14.3 to

3

16.6 g/d for women (Health Canada and Statistics Canada, 2004). Despite current

recommendations and growing awareness surrounding the importance of fibre

consumption, reported low intake levels suggest a need for fibre fortification of

foods.

Hydrocolloids are commonly used ingredients by the industry for their ability

to impart technological functionality, such as colloidal stability, texture, ice

recrystallization inhibition, gelation and viscosity, in a range of food systems.

Furthermore, the ability of hydrocolloids to act as dietary fibre makes them

functional food ingredients as well, allowing them to deliver at least one of the

aforementioned health benefits. Because of this they are integral components of

human diets as they can define both the sensory and nutritional qualities of foods

(Gidley, 2013). Therefore, selecting the most appropriate ingredient for food

fortification means maximizing nutritional quality without compromising

organoleptic properties of the product.

Viscosity is a physicochemical property of foods that can be controlled

through the addition of certain hydrocolloids, such as guar gum and xanthan gum.

Moreover, it is purported to be associated with some of the health benefits that arise

from consumption of fibre rich foods, which includes attenuation in postprandial

glycaemia (Brownlee, 2011). Fibre intake has been reported to be associated with

improved glucose/insulin metabolism in T2D patients (Jenkins, Jenkins,

Zdravkovic, Wursch, and Vuksan, 2002) and observational studies indicate that diets

high in fibre have an overall protective effect against T2D (Barclay, Petocz,

McMillan-Price, Flood, Prvan, Mitchell, and Brand-Miller, 2008; Kendall, Esfahani,

4

and Jenkins, 2010). The ability of viscous dietary fibres to reduce plasma glucose

levels in humans has been well reported over the past couple of decades (Jenkins et

al., 1977). However, reductions in plasma glucose concentrations achieved by

increasing viscosity may be due to one of several events (Dikeman & Fahey Jr.,

2006). Carbohydrate assimilation, which will be discussed in detail in Chapter 2,

begins inside the mouth. As the bolus reaches the stomach an increase in viscosity of

the gastric contents may slow gastric emptying rates. Furthermore, a thickening

inside the small intestine may decrease contact between the small intestinal contents

and digestive enzymes. Finally at the mucosal surface, ingestion of viscous fibres

may lead to a thickening of the unstirred water layer, thereby slowing the diffusion

rate of glucose (Edwards, Johnson, and Read, 1988; Eastwood and Morris, 1992;

Mälkki, 2001). Despite the growing evidence and continual research over the past

couple of decades, the exact mechanism(s) by which dietary fibres are able to

modulate the glycaemic response in humans remains to be elucidated.

5

1.2. Overall Objectives

In our earlier work we have shown how despite increasing solution viscosity,

through the addition of various types of dietary fibre, substantial reductions in

digesta viscosity were observed for all solutions following a simulated digestion

protocol (Fabek, 2011). Furthermore, the ability of fibres to resist loss of viscosity

varied, with some exhibiting total loss while others were able to retain a significant

amount. However, the design protocol did not include adding fibre to a food matrix

containing other nutrients such as available carbohydrates and protein. Moreover, the

link between digesta viscosity and physiological functionality was not examined.

The overall objective of this study is to investigate the structure-function

relationship of viscous soluble fibres, by focusing on the effect(s) they may have on

reducing glucose levels, in vitro. The milestones of this work are:

1) To formulate solutions containing protein and starch with guar gum,

xanthan gum, soluble flaxseed gum, and soy soluble polysaccharide

(SSPS) at concentrations that allow for similar solution viscosities

2) To analyze the rheological behaviour of fibre-fortified solutions

throughout in vitro digestion

3) To determine the amount of intermediate starch hydrolysis products and

glucose released in the simulated digesta

4) To examine the effects that fibre addition may have on glucose

diffusivity inside a dialysis system

6

5) To isolate starch granules at various times of simulated small intestinal

digestion for analysis using light scattering, scanning electron

microscopy and light microscopy

7

2. Literature Review

This review will primarily focus on the relationship between dietary fibre,

specifically soluble dietary fibre, and the ability to generate glycemic reductions.

Initially, an overview of the biomechanics of the gastrointestinal tract will be

presented. This will be followed by a description of nutrient absorption, with the focus

being on carbohydrate metabolism throughout the salivary, gastric, and small

intestinal phases. Subsequently, recent advances in the design protocols of in vitro

digestion methods will be reviewed, with a focus on studies exploring starch

digestion. Information on dietary fibre, including the definition, classification and

postulated health benefits of its subsequent consumption will be outlined. Finally, the

structure, physicochemical properties, and uses of guar gum, xanthan gum, flaxseed

gum, and soy soluble polysaccharide will be presented.

2.1. Digestion of foods

2.1.1. Gastrointestinal physiology

The gastrointestinal tract (GIT) is the largest endocrine organ in the body and

stretching it would create a long muscular tube close to 30 ft in length. The objective

of the GIT is digestion of foods, absorption of nutrients, and excretion of waste

products. Surprisingly, the process of digestion begins before food is even placed

inside the mouth, this is known as the cephalic phase of digestion. Chemoreceptors

and mechanoreceptors located in the buccal and nasal cavities are stimulated by the

mere thought or smell of food. Subsequently, nerve impulses are transmitted to the

stomach via vagal efferent fibers, which initiates gastric acid secretion in preparation

8

for food entering the GIT. The cephalic phase is responsible for nearly 30% of the

total response to a meal (Johnson, 2014).

As we ingest food, the process of chewing reduces the size of ingested food

particles and mixes the food with saliva thereby lubricating the bolus and exposing it

to digestive enzymes, such as salivary amylase. The bolus is then propelled in a

proximal-to-distal manner through the pharynx via peristaltic contractions, which are

rhythmic wave-like contractions of the pharyngeal musculature. The bolus is then

further propelled through the esophagus in a similar manner of muscular contractions,

where inner circular muscle fibers squeeze the food and the outer longitudinal muscle

layer physically pushes the food toward the stomach. Often a secondary peristaltic

event takes place, which aids in emptying the esophagus or if gastric contents reflux

back into the esophagus (Johnson, 2014).

The stomach is separated into three distinct anatomical divisions, the fundus,

the corpus (or body), and the antrum. The fundus acts as a reservoir for ingested

foods. It has been shown that one of the tasks required by the stomach is to act as a

temporary storage compartment, controlling the release of gastric contents into the

duodenum (Fengua and Singh, 2010). The body is responsible for propelling gastric

contents toward the pylorus. Lastly, the antrum is characterized by high intensity

mechanical mixing allowing for chyme to be adequately mixed with gastric secretions

(Chang, Sitrin, and Black, 1996). As previously mentioned, the stomach begins

preparation before the bolus reaches the lower esophageal sphincter. Gastrin, a

hormone secreted by the G cells of the stomach lining, stimulates gastric glands in the

oxyntic gland area along with histamine and acetylcholine (Johnson, 1991). These

9

gastric glands contain two types of cells: parietal cells and chief cells. The former

cells secrete intrinsic factor, a mucoprotein critical for the absorption of vitamin B12,

and hydrochloric acid responsible for creating a low pH environment. The latter cells

are responsible for secreting pepsinogen, the enzyme precursor to pepsin that initiates

protein digestion, and gastric lipase responsible for lipid digestion. Furthermore, the

gastric mucosa contains epithelial cells that subsequently secrete soluble mucus

leading to the lubrication of chyme as well as insoluble mucus that protects the

mucosa from damage (Johnson, 2014). Overall, gastric acid secretion accounts for

more than 50% of the total response to a meal.

As mentioned previously, the task of the orad region of the stomach is to

accommodate ingested material. It contains weak contractions and thin musculature to

permit stretching. In the fasted state, stomach volume is approximately 200 mL and

this increases to more than 1.7 L in the fed state allowing for dimensions of

approximately 10 cm wide and 30 cm long to be reported (Ferrua and Singh, 2010).

Once chyme enters the body of the stomach peristaltic contractions, with a frequency

of three contractions per minute and duration of 2 to 20 seconds per contraction, mix

and propel gastric contents toward the gastroduodenal junction, or pylorus. The

velocity of the peristaltic wave, which is controlled by fluctuations in membrane

potential, increases as it reaches the pylorus allowing it to eventually overtake the

gastric contents. As this occurs, everything but the smallest food particles is pushed

back, a process known as retropulsion. Ultimately, this provides sufficient mixing

leading to a breakdown of food particles to a small enough size, less than 1 mm3,

thereby permitting passage through the pylorus and into the small intestine. An earlier

10

study done on examining the relationship between gastric motility patterns and gastric

emptying rates of solids and liquids in humans demonstrates how liquid meals exit the

stomach at a much faster rate than do solid meals, as can be seen in Figure 2-1. The

authors attribute their findings to the requirement of the solid component to be

reduced to small particles before being emptied into the small intestine.

Figure 2-1. A comparison of solid and liquid components remaining in the stomach

after ingestion of a meal (Adapted from Camilleri, Malagelada, and Brown, 1985).

The small intestine is divided into three sections, the duodenum is the section

distal to the pylorus of the stomach, the jejunum is the middle portion, and the ileum

is the most distal section that connects to the large intestine via the ileocecal valve.

Altogether, they constitute the longest portion of the GIT and the small intestine is the

11

site where most of the nutrient absorption takes place, which will be discussed in

greater detail in Section 2.1.2. As food particles pass through the pylorus reaching the

duodenum, the pancreas is stimulated resulting in alkaline and enzymatic secretions.

The former is stimulated by secretin, also known as “nature’s antacid”, which is

released from S cells in response to gastric acid and increasing levels of long-chain

fatty acids. The latter is stimulated by cholecystokinin (CCK), arriving from I cells in

response to products of protein (amino acids) and fat (fatty acids) digestion.

Neutralization of food particles occurs rapidly in the duodenum and after

approximately 30 minutes enzyme secretion reaches nearly 80% completion and

continues until the stomach is empty. Pancreatic secretions are produced at a rate of

0.2 – 0.3 mL/min and can reach values of up to 4 – 4.5 mL/min, when stimulated

(Chang et al., 1996). In the case of lipid digestion, bile, which is secreted from the

liver and is comprised of bile acids, phospholipids, cholesterol, and pigments, is

released from the gall bladder via the common bile duct to aid in emulsification of the

lipid compounds (Johnson, 2014).

The small intestine is innervated by both the autonomic and enteric nervous

systems, which are together responsible for the contractions inside the small intestine.

In particular, two layers of smooth muscles control the contractile activity, an outer

longitudinal layer and a thicker inner circular muscle layer. Both layers are found

abundant throughout the small intestine and their thickness has been reported to

decrease distally towards the aforementioned ileocecal junction (Johnson, 2014).

Similar to the stomach, smooth muscles of the small intestine undergo cyclic

depolarizations and repolarizations inducing what is known as slow wave activity.

12

Unlike the contractions in the stomach, however, the contractions here are initiated by

spike potentials, which are rapid depolarizations of the cell membrane that occur

during the depolarization phase of the preceding slow wave. The movements of the

small intestinal walls are divided into 2 types of contractile events: segmentation and

peristalsis. Segmentation occurs when slow waves are induced in the absence of spike

potentials. They are localized concentric contractions, due to the fact that the

contraction is neither preceded nor followed by further contractions, which are 1 to 2

cm in length and that act to physically squeeze the chyme and mix it with digestive

enzymes. However, if contractions at adjacent sites occur, then contents are propelled

in an analward direction at a velocity of 0.5 to 2 cm/s, this is known as peristalsis

(Guyton & Hall, 2010; Tharakan, Norton, and Bakalis, 2010; Johnson, 2014). The

continuous contraction and relaxation of the circular and longitudinal muscles push

the intestinal contents toward the ileocecal junction so that the remaining material

may pass into the large intestine. The mechanical design of the small intestine makes

it a vital organ for nutrient absorption.

2.1.2. Carbohydrate metabolism

Carbohydrates are an important energy source for the body, providing

approximately 4 kcal/g, and are the primary fuel source for the brain. Currently,

Health Canada advocates that the acceptable macronutrient distribution range, as the

percentage of total calories being derived from carbohydrates, should fall between 45

– 65% of the total diet (Health Canada, 2011). Dietary carbohydrates can be further

divided into sugars, starch and fibre; the latter term will be discussed in detail in the

next chapter. Starch, a storage polysaccharide in plants, is the main energy source in

13

human food (Hasjim, Lavau, Gidley, and Gilbert, 2010) and is comprised of two

molecules, amylose and amylopectin. Amylose is the smaller of the two, with a

molecular weight of approximately 106 Da and is primarily linear with few longer

chain branches. Amylopectin on the other hand is a larger, highly branched molecule

with a molecular weight of approximately 108 Da. Both consist solely of D-glucose

connected by α-1-4 linkages, with amylopectin containing α-1-6 linkages at the

branched points. Both polymers are α-glucans and, therefore, their subsequent

digestion is initiated in the mouth.

Saliva, the secretion of which is almost completely controlled by the nervous

system, functions by protecting the mouth, lubricating the bolus, and in the case of

starchy foods initiating the process of digestion. Ptaylin is the α-amylase enzyme

found in the mouth and begins to cleave the α-1,4-glycosidic bonds at random

locations along the chain of the entering carbohydrate moiety. The swallowed bolus

then travels down through the esophagus into the stomach where the acidic

environment deactivates α-amylase and enzyme activity ceases altogether. Food

particles are further propelled through the pylorus, as described in Section 2.1.1.,

arriving in the duodenum, which is where the majority of starch digestion begins.

Digestion within the small intestine occurs both inside the lumen and at the

surface of epithelial cells, through the action of enzymes embedded along the brush

border. Small intestinal luminal digestion begins when chyme is sufficiently mixed

with pancreatic secretions containing amylase enzymes that are secreted from the

pancreatic duct. These enzymes attack the interior α-1,4-bonds, producing maltose and

the trisaccharide maltotriose as well as α-limit dextrins, which are branched

14

oligosaccharides of glucose containing 5 to 10 glucose residues and are formed from

the hydrolysis of amylopectin. Through segmentation these luminal digestion products

are brought in contact with brush border carbohydrases. Maltase hydrolyzes maltose

into glucose, sucrase cleaves sucrose to yield glucose and fructose, and lactase breaks

down lactose into glucose and galactose; individuals with insufficient amounts of

lactase suffer from lactose intolerance. Finally, glucoamylase liberates glucose

residues starting from the non-reducing ends until a 1-6 branch point is reached,

which is subsequently hydrolyzed by isomaltase. Unless there is a predisposing

medical condition, sugar assimilation in humans is complete in the proximal jejunum.

There seems to be minimal control in humans over how much free sugar is absorbed,

and it has been estimated that hexoses equivalent to 22 pounds of sucrose can be

absorbed per day (Johnson, 2014). However, there are reports of a negative feedback

mechanism by which chemo- and osmoreceptors in the proximal small intestine

regulate the emptying of gastric contents. To exemplify this, a 750 mL isotonic citrate

solution that is placed in a human stomach passes through the pylorus in 20 min

whereas a solution with the same volume containing sucrose is delivered at a much

slower rate (Johnson, 2014).

2.1.3. Glucose absorption

In order for the body to be able to utilize carbohydrates as a fuel source, the

monosaccharides must first get absorbed into the bloodstream. Despite the presence of

aqueous channels between enterocytes, glucose is too large of a molecule to pass

through via simple diffusion; therefore, both passive and active transport are involved.

In humans, there are three major routes used to allow for monosaccharide absorption.

15

Fructose is transported from the intestinal lumen via glucose transporter 5 (GLUT-5)

and exits the cell using a different channel, GLUT-2. Glucose and galactose enter the

blood using the same channel, GLUT-2, however unlike fructose, which is absorbed

by facilitated diffusion alone, they are absorbed by active transport via an Na+-

dependent transport system, SGLT-1. The carrier binds 2 Na+ molecules for every

glucose or galactose molecule being transported, as illustrated in Figure 2-2. The

reaction is ATP-driven and maintains a Na+ gradient that favours glucose entry.

Glucose exits the cytosol using GLUT-2, located at the basolateral membrane, and

travels to the liver via the portal vein, which is responsible for moderating the

distribution of glucose throughout the body.

Figure 2-2. GLUT2 model showing intestinal aborption of glucose following a meal

(Adapted from Kellett and Brot-Laroche, 2005).

16

The target plasma glucose value is 7.8 mmol/L or less, anything between 7.8

and 10 is considered pre-diabetes or Impaired Glucose Tolerance (IGT), 10 and

beyond classifies diabetes (Canadian Diabetes Association, 2009). Although insulin is

responsible for stimulating liver and muscle cells to utilize glucose from the blood,

inadequate production may lead to type I diabetes, and if cells are resistant to the

insulin that is produced, then type II diabetes may develop. In either case,

hyperglycemia is the result and in the latter hyperinsulinemia may also occur due to

the continuous production of insulin by the pancreas. As mentioned in Chapter 1, the

occurrence of T2D is on the rise, which is why understanding the mechanics of

carbohydrate digestion is crucial. However, studying digestion behaviour, in vivo, can

be both costly and time consuming with many ethical considerations and restrictions

that cannot be overlooked. As a result, developing in vitro techniques has become an

integral design strategy for research that is focused on trying to better understand

human digestive health.

2.1.4. Review of in vitro digestion

Simulated digestion protocols more often than not incorporate the oral, gastric

and small intestinal phases. Occasionally, depending on the nature of the research, a

large intestinal fermentation step is included (Minekus, Alminger, Alvito, Balance,

Bohn, Bourlieu, et al., 2014). Regardless of the in vitro digestion method that is

employed, the strategy is to mimic physiological conditions and therefore the design

takes into account pH, incubation times, digestive enzyme concentrations and

agitation speeds. These principles date as far back as 1969 when Southgate

incorporated amyloglucosidase and pullulanase to hydrolyze starch (Southgate, 1969).

17

A 2011 study, which surveyed more than 80 publications whose research focused on

in vitro digestion models for foods, reported that the majority (45%) were based on

plant-based foods such as starch, tea and bread, followed by meats (18%), and

emulsions (9%) (Jur, Lim, Decker, and McClements, 2011). Since the time of

Southgate, two types of digestion models have been introduced to allow researchers to

study the gastrointestinal behaviour of foods and pharmaceuticals; those are static

models and dynamic models.

Static digestion models receive their name from the fact that the concentrations

of enzymes, bile acids, buffers (pH) and salts and their subsequent ratios to meal

components, remain the same during each stage of digestion. Nonetheless, they

comprise the majority of models that are used in studying the digestibility and

bioavailability of mycotoxins, soil contaminants, pharmaceuticals, and macronutrients

such as proteins, lipids and carbohydrates (Guerra, Etienne-Mesmin, Livrelli, Denis,

Blanquet-Diot, and Alric, 2012). The design includes a brief oral phase, usually 5 min,

to allow for mechanical breakdown and to initiate amylolysis. Subsequently, a gastric

step (1 – 2 h) continues the breakdown process and mixes the substrate with simulated

gastric fluid containing hydrochloric acid and pepsin. Finally, the simulated intestinal

fluid of the intestinal phase neutralizes the mixture and supplies a variety of different

enzymes as well as bile to allow for complete digestion of protein, fat and

carbohydrate. The common theme amongst them is that from the oral step to the small

intestinal stage, all enzyme concentrations, pH values, mechanical stresses and

incubation times are fixed. Although they provide simplicity and a high throughput,

static models lack a physiological relevance to in vivo conditions, which is a challenge

18

for any in vitro digestion protocol. The mechanics of the GIT are highly dynamic

throughout all stages of digestion. As was illustrated in Section 2.1.1, the digestive

tract is a highly complex system that incorporates a variety of control/feedback

mechanisms, including neurological, hormonal, and even the nutrients themselves can

regulate different steps of digestion. For example, in the oral step hydration-

lubrication depends on the meal type, whether it is solid or liquid. In the gastric stage,

only small particles may pass through the pylorus, indicating that not everything is

introduced to the duodenum at the same time. Moreover, the pH in the stomach during

the fasted state is significantly different than after consumption of a meal. In the small

intestine, the lack of removal of digestive products is a major drawback to using static

models, which has been reported to lead to product inhibition of digestive enzymes

(Minekus et al., 2014). Additionally, there are both intra- and inter-personal

variability where age, physical constitution, mental status, and general eating habits

all factor in on how food is metabolised. These are parameters that are nearly

impossible to control and furthermore it is difficult to compare the results of one static

model to another due to the heterogeneity in their design protocols (pH, incubation

times, enzymes used in the study and their concentrations). These drawbacks have led

researchers to create novel ways by which human digestion may be studied, through

the introduction of dynamic digestion models.

The dynamic gastric model (DGM) simulates events taking place inside the

stomach. The design includes two compartments, the body and the antrum, the

schematic is illustrated in Figure 2-3. The body is essentially a cone that can

accommodate up to 800 mL of material, which is placed inside a water bath. Changes

19

in water pressure act to mimic peristaltic actions in the stomach (three gentle

contractions are produced per minute). It is inside the body where digestive secretions

are released. The secretion of enzymes is achieved using a peristaltic pump, which is

connected to a computer that calculates the enzyme concentrations based on the

volume of food inside the machine. The pH is monitored using a pH catheter placed

directly inside the body. The antrum is responsible for the mixing and subsequent

breakdown of food material. The grinding of the antrum occurs by the sliding of two

stainless steel cylinders controlled by a piston, which act to break down the food

particles to allow them to pass through an elastic annulus, a small opening on the

bottom representing the pylorus. The unique design of the DGM allows it to mimic

retropulsion, where only smaller particles are able to leave the stomach to allow for

subsequent analysis and larger ones are refluxed back into the top chamber (body) to

allow for further mixing with gastric secretions (Chessa, Huatan, Levina, Mehta,

Ferrizzi, and Rajabi-Siahboomi, 2014; Wickham, Faulks, Mann, and Mandalari, 2012;

Vardakou, Mercuri, Barker, Craig, Faulks, and Wickham, 2011).

20

Figure 2-3. Schematic diagram of the dynamic gastric model. Numbers illustrate

passage of meal contents from the (1) the main body to the (2) the antrum to allow for

processing using (3) the piston and barrel to allow for (4) analysis (Adapted from

Vardakou et al., 2011).

Other dynamic models exist that are compartmental, mimicking the conditions

inside the stomach, and both the small and large intestines. Similar to static models,

they are able to reproduce physiological temperature, pH, and enzymes additions.

Unlike static models, however, they are able to do so using computer systems, where

the addition of secretions and mechanical events such as gastric emptying, peristaltic

mixing and transport, and passive absorption of water are controlled. An example of

this is the TNO gastro-Intestinal Model 1 (TIM-1). The model is comprised of the

21

stomach and the three parts of the small intestine (duodenum, jejunum and ileum) and

is considered to be the closest simulation to in vivo digestion (Guerra et al., 2012).

Despite continuous advances and technological efforts to improve in vitro methods

there are always limitations in modelling human digestion. As mentioned, static

models continue to be the preferred method; however, due to the heterogeneity

amongst the different models, there is a need for standardisation of in vitro digestion

protocols.

The requirement for standardisation of in vitro methods is the result of the

range in values for incubation times, buffer constituents and enzyme concentrations

that are employed and reported in literature. Hollebeeck, Borlon, Schneider,

Larondelle, and Rogez (2013) have reported that incubation times can range from 5

min to 15 min in the salivary step, from 30 min to 180 min for the gastric step, and

from 60 min to 360 min for the small intestinal step. Moreover, pH levels reported in

recent literature vary from 5.0 to 6.9 in the salivary step, from 1.1 to 2.8 in the gastric

step, and from 6.3 to 7.8 for the small intestinal step. Finally, the composition and

concentration of enzymes employed in different studies and the lack of reported

enzyme activity makes it difficult to compare studies that use static models

(Hollebeeck, et al., 2013). An earlier study demonstrated the necessity of

incorporating pepsin in the simulated gastric step and the effect that varying pH of the

simulated duodenal step has on pancreatin activity (Woolnough, Monro, Brennan, and

Bird, 2008). As a result, studies have been done to explore the effect that three

continuous factors (pH, incubation times and enzyme concentration) have on

macronutrient digestibility, in order to ascertain a standardised protocol (Minekus et

22

al., 2014; Hollebeeck et al., 2013). One of these suggested design protocols is

illustrated in Figure 2-4.

Figure 2-4. Proposed flow diagram for a standardised in vitro digestion method

(Adapted from Minekus et al., 2014)

Although there is variation in the proposed parameters among studies seeking

standardisation, the focus to standardise in vitro protocols will assuredly allow for

more accurate inter-study comparisons and overall consensus amongst researchers.

23

2.1.5. In vitro digestion methods related to starch digestion

As mentioned in Section 2.1.4, digestion methods have been around since

1969, when Southgate used pullulanase and amyloglucosidase to hydrolyze starch.

Since then, methods for modelling amylolysis of starch have been improved upon and

are widespread throughout literature. However, the Englyst method, which employs

controlled enzymic hydrolysis to measure total starch, rapidly digestible starch, slowly

digestible starch and resistant starch fractions in foods (Englyst, Kingman, and

Cummings, 1992), has become the pillar of in vitro models investigating starch

digestion and is heavily cited among nutrition and food science journals (Edwards,

Warren, Milligan, Butterworth, and Ellis, 2014). In fact, both the Englyst and

Southgate procedures have become two of the most widely used enzymatic chemical

methods for dietary fibre analysis, along with the AOAC enzymatic gravimetric

methods (Tungland and Meyer, 2002). Mimicking the in vivo circumstance is a

challenging task, which is made even more challenging when attempting to use in

vitro methods to analyse the mechanics and products of starch digestion. The

challenge is associated with controlling experimental conditions, as they are affected

by a variety of parameters that are intrinsic to not only the biochemical and physical

processes/conditions of the body but also the structural organization of different starch

granules. Consequently, research on in vitro starch digestion is of immense

importance to biochemists, physicists, nutritionists and food scientists as well as

researchers exploring the requirements for a stadardised in vitro design protocol.

Studies focused on exploring the need for standardised in vitro digestion

methods use starch as a substrate to determine the fixed conditions of the salivary

24

step. Minekus et al. (2014) used soluble potato starch to assess amylase activity and

concluded that a 2 min salivary stage at pH 6.8 with α-amylase concentration of 75 U

mL-1

should be used as a standard. In contrast, Hollebeeck et al. (2013) used surface

methodology to determine the incubation time, pH, and digestive enzyme

concentrations that will allow for 5% salivary starch hydrolysis, mimicking the in vivo

circumstance (Guyton and Hall, 2010). The authors of this paper concluded that a 5

min salivary step at pH 6.9 and amylase activity of 1.3 U mL-1

are sufficient to allow

for 5% starch hydrolysis. Despite varied results, this demonstrates the continued

importance of using starch as a substrate for in vitro digestion models. In addition,

studies have explored enzyme kinetics by modelling amylolysis of starch-containing

foods. Edwards et al. (2013) used the Log of Slope (LOS) analysis of digestibility

curves in order to quantify nutritionally important starch fractions. The researchers

purified starch from chickpeas and durum wheat and analysed in vitro digestibility

following the addition of porcine pancreatic α-amylase. Subsequently, LOS analysis

was applied to the digestibility curves in order to determine first order kinetic data

using a first-order equation (equation 1-1).

Equation 1-1. First order equation.

𝐶𝑡 = 𝐶∞ (1 − 𝑒−𝑘𝑡 )

(where Ct = concentration of product at given time (t), C∞ = concentration of product

at the end of the reaction and k = digestibility rate constant).

By fitting the LOS analysis to the experimentally-derived digestibility curves they

were able to demonstrate that amylolysis occurs in both single-phase and two-phases,

25

depending on the particle size. The authors concluded that C∞ and k, and LOS plots in

general, are valid predictors of the release of hydrolysed starch products arising from

amylolysis (Edwards et al., 2013). In a different study, researchers exploring the

mechanism of action of amylase enzymes were able to successfully label porcine

pancreatic α-amylase using two different dyes. Subsequently, they observed the

labelled α-amylase using a confocal laser scanning microscope of unhydrolysed and

hydrolysed starch. The results of the study demonstrate a preferential binding of the

enzyme to more susceptible parts of the granule, such as the hilum where there is less

organization, molecular order and greater overall porosity. As the enzyme was bound

to the granule, more binding sites were exposed, which led to further digestion of the

starch granules (Dhital, Warren, Zhang, and Gidley, 2014). A 2010 study compared

the effect of an in vitro digestion method to in vivo (pig) digestion. Comparisons were

made on the basis of changes in the morphology of isolated starch granules of in vivo

and in vitro digesta. The authors observed that the starch granules of the in vivo

digesta were almost completely hydrolysed in the upper part of the small intestine of

the pig. This, however, was not observed in the granules isolated from the in vitro

digesta. Moreover, they observed heterogeneity in the attack of enzymes on the

granules in the in vivo digesta, whereas changes in morphology of granules isolated

from the in vitro digesta appeared to be even across all granules, as shown in Figures

2-5 (a) and (b). The authors concluded that the conditions present during in vivo

digestion were not accurately produced using their in vitro protocol (Hasjim et al.,

2010).

26

Figure 2-5. Morphology of (a) raw maize starch granules isolated from in vivo digesta

and (b) starch granules isolated from in vitro digesta (Adapted from Hasjim et al.,

2010).

In a different study, researchers used an in vitro model to assess starch digestion

kinetics and particle size distributions. The authors concluded that in vitro starch

digestion followed first-order kinetics and that the size of starch granules affected the

rate of digestion, where smaller particle sizes allowed for more rapid digestion due to

increased surface area. Moreover, starch properties such as architecture, crystalline

pattern, degree of polymerisation, and the presence of surface pores or channels also

influence digestibility (Mahasukhonthachat, Sopade, and Gidley, 2010).

27

As mentioned in Section 2.1.2, starch is the primary fuel for the body. For

decades, the nutritional importance of this energy-providing carbohydrate has sparked

interest of nutritionists and food scientists that are exploring ways to reduce diet-

related diseases. Although a plethora of human studies have been conducted,

including the 1981 study that resulted in the coining of the term ‘glycemic index’

(Jenkins, Wolever, Taylor, Barker, Fielden, Baldwin, Bowling, Newman, Jenkins, and

Goff 1981), researchers employ in vitro methods to further our understanding of the

mechanics of starch digestion and factors influencing metabolic effects. Hasjim et al.

(2010) demonstrated that in vitro digestion of a pig diet containing raw normal maize

starch liberated nearly half of the starch as glucose following a 5 h incubation period

and 98% of the starch had broken down into glucose following 25 h of in vitro

digestion. A more recent study was carried out that exposed a variety of starches (sour

polvilho, sweet polvilho, tapioca, and cassava flour) to the aforementioned TIM-1

dynamic model. The purpose of the study was to determine a starch that allows for

slow glucose release, which may be considered ideal for consumption by patients

suffering from hepatic glycogen storage disease (GSD). The authors concluded that

sweet polvilho allowed for a 15% reduction in the rate of glucose release in

comparison to uncooked cornstarch, which has been a treatment option for GSD

patients since the 1980’s (Nalin, Koen, Weinstein, de Souza, Perry, van Wandelen,

van Rijn, Smit, Schwartz, and Derks, 2014).

Due to the processing of starch, the quality of carbohydrates is often reduced

and this is concomitant with a higher glycemic index. Subsequently, a study was

carried out to characterize starch fractions of foods as rapidly digestible starch (RDS),

28

slowly digestible starch (SDS), and resistant starch (RS) by measuring the amount of

glucose using a standardized in vitro technique. The researchers exposed spaghetti,

pearled barley, corn flakes and white bread to a two-stage in vitro digestion model.

Samples were taken after 20 min (RDS component) and 120 min (SDS component)

and were subsequently analysed for sugar concentrations using high-performance

liquid chromatography. Furthermore, the study included a clinical component through

the recruitment of eight volunteers. Participants were asked to consume four test foods

and finger-pick samples were taken at various times to measure plasma glucose. The

results demonstrated an inverse relationship between the amount of RAG measured in

the test foods and blood glucose levels. Corn flakes and white bread showed the

highest values of RAG and as a result elicited significantly higher in vivo glycaemic

responses. They concluded that RAG intake is a major determinant of the glycaemic

response of a food (Englyst, Englyst, Hudson, Cole, and Cummings, 1999). A recent

follow up to this study was done to determine the relationship between cornstarch

digestibility and glycemic response and subjective appetite in young men. The study

included starches with a wide range in RDS and RS, which were measured using the

Englyst method (Englyst et al., 1999). Whole-grain (10 g RDS), regular cornstarch

(13 g RDS), high-amylose starch (19 g RDS) and maltodextrin (41 g RDS) were

employed in the study. Subsequently, the starches were added to tomato soup and the

subjects were asked to fill out questionnaires to rate subjective appetite after

consumption of an ad libitum pizza meal that was offered 30 min and 120 min after

the test meals. Blood samples were also taken at various times throughout the study.

The results showed that the maltodextrin and whole-grain treatments led to the highest

29

and lowest blood glucose levels, respectively. Food intake, on the other hand, was

associated with higher levels of RS, which they believe is attributed by satiety signals

in the body. The authors concluded that the in vitro classification of starch

digestibility using the aforementioned Englyst method was able to predict the

glycemic response and food intake levels in the participants (Anderson, Cho,

Akhavan, Mollard, Luhovyy, and Finocchiaro, 2010). The unremitting research on

starch digestion has led to advances in in vitro digestion techniques and has generated

a greater understanding of the health implications associated with consumption of

foods high in digestible carbohydrates. However, much of the focus has been shifted

to understanding the biomechanics of the GIT and functional properties as related to

consumption of indigestible carbohydrates, or dietary fibre.

2.2. Dietary Fibre

2.2.1. Dietary fibre definition

Eben Hipsley was accredited for coining the term ‘dietary fibre’ in 1953, when he

observed that populations who consume fibre-rich foods also reported lower incidence

rates of pregnancy toxaemia (Brownlee, 2011; Hipsley, 1953). Hippocrates in 430 BC

described how coarse wheat promoted a laxative effect, which was not evident in

refined wheat, and in the early 1920’s, J.H. Kellogg related increased stool weight and

laxation to bran consumption, which sparked interest in dietary fibre research (Slavin,

2013). This notion of having a physiological benefit arise as a direct result of fibre

consumption remains to be the backbone of the current definition of dietary fibre.

30

There are various contortions and arguments surrounding the exact definition and

classification of fibre, depending on the scientific body. However, the 2009 meeting

of Codex agreed that dietary fibre is a carbohydrate with 10 or more monomeric units

(DP≥10), is of plant origin and escapes digestion in the human upper GIT (Codex,

2009; Institute of Medicine, 2008; EFSA, 2010; ANZFA, 2000; and Health Canada,

1985). In addition, Health Canada’s definition necessitates that it must deliver at least

one physiological benefit to the consumer. However, because the definition excludes

oligosaccharides (DP < 10), resistant starches (RS1, RS2, RS3 and RS4), and requires

that all dietary fibre must be of plant origin, it makes Health Canada’s definition very

restrictive (Jones, Lineback, and Levine, 2006). Accordingly, a proposal for a new

definition was put forth in 2010 that reads, “Dietary fibre consists of naturally

occurring edible carbohydrates (DP>2) of plant origin that are not digested and

absorbed by the small intestine and includes accepted novel dietary fibres” (Health

Canada, 2010). As discussed in Chapter 1, dietary fibre consumption in Canada falls

far below the recommended levels put forth by Health Canada. Therefore, accepting a

broader definition of dietary fibre will place fewer restrictions on the food industry

that is afforded the task of formulating foods with novel fibre ingredients. As a result,

this will also help consumers by increasing the availability and accessibility to fibre-

rich foods in today’s market, thereby delivering the proposed health benefits. Many

stakeholders, however, asked for more unambiguous explanations of what constitutes

a physiological effect, which is synonymous with Health Canada’s current definition

of dietary fibre. In response, the benefits include one of: improved laxation and

increased fecal bulk, reduced level of triglycerides and/or LDL cholesterol, reductions

31

in post-prandial plasma glucose and/or insulin levels and/or supply of energy-yielding

metabolites and short-chain fatty acids (SCFA’s) through colonic fermentation. A

product supplemented with a novel dietary fibre and without a prior history of food

use must therefore demonstrate both safety and an ensuing physiological benefit that

will arise from its subsequent consumption before it can be placed on the market

(Phillips, 2013).

2.2.2. Health benefits of dietary fibre

Despite the fact that the caloric value of dietary fibre is approximately 2

kcal/g, it remains a vital nutrient in our everyday diet due to its ability to promote

physiological benefits throughout the body. As mentioned in Section 2.2.1, dietary

fibre, according to the definition, has to exert at least one of the benefits that are listed

by Health Canada, but not all. Therefore, it is important to highlight that the resulting

physiological effect is dependent on the type of fibre consumed, that is, not all fibres

are alike (Slavin and Jacobs, 2010). Although they are all indigestible, structural

dissimilarities, such as number and type of monomeric units, types of linkages present

between monomers, and presence of branch points and charged groups along the

backbone (Tungland and Meyer, 2002) create heterogeneity in dietary fibre

functionality. Subsequently, water solubility may help to further distinguish and

classify fibre types. Water soluble fibres, such as gums and pectins, rely on their

ability to thicken when they come in contact with gastric and intestinal secretions and

the ensuing viscosity is related to health benefits along the upper GIT (Dikeman et al.,

2006). On the other hand, water insoluble fibres such as cellulose, hemicellulose and

32

lignin are not affected as they pass through the upper GIT and physiological

functionality is solely associated with colonic fermentation (Brownlee, 2011).

The human large intestine is heavily colonized with up to 1000 different

species of bacteria and can reach numbers of 1011

to 1012

anaerobic bacteria for every

gram of intestinal content (Roberfroid, Van Loo, and Gibson 1998). Colonic bacteria

are able to hydrolyze indigestible (insoluble) carbohydrates to produce gases

(hydrogen, carbon dioxide and methane) as well as SCFA’s (acetate, propionate and

butyrate). Increased levels of SCFA’s have been shown to be vital to the health of the

colonic epithelium (Topping and Clifton, 2001). In addition to stimulating colonic

fermentation, insoluble fibres increase fecal mass thereby having a stool bulking

effect as well. This increased weight reduces transit time in the large intestine and

may help prevent or relieve constipation (Slavin, 2013). Inulin, oligofructose, and

fructo-oligosaccharides demonstrate a prebiotic effect by increasing the level of

bifidobacteria at low levels of consumption. This was also demonstrated in studies

working with acacia gum and polydextrose, which have all led to increases in levels of

bifidobacteria (Jie, Bang-Yao, Ming-Jie, Hai-wei, Zu-kang, Ting-song, and Craig,

2000; Hengst, Ptok, Roessler, Fechner, and Jahreis, 2008). The results of a clinical

trial conducted on 40 female subjects showed that wheat dextrin supplementation for

fourteen days resulted in higher levels of bacteroides and depletion of pathogenic

bacteria (Slavin, 2013). Earlier studies have also shown that consumption of alginate,

chitosan and inulin all lead to reductions in harmful microfloral metabolties (Grasten,

Liukkonen, Chrevatidis, El-Nezami, Poutanen, and Mykkanen, 2003). In addition to

dietary fibre having a profound influence on the colonic microflora there are also

33

reports that consumption of fibre-rich foods may increase gastrointestinal immunity.

One of the roles of the colonic mucosa is to assess whether luminal contents contain

any toxic compounds and this is achieved through the gut-associated lymphoid tissue.

Samples of colonic digesta are taken up by M cells situated along the large intestinal

epithelium. Antigens present in the samples may then elicit an immune response via

the systemic immune system. Animal studies have shown that highly fermentable

fibres (fructooligosaccharides and gum Arabic) have led to increased T-cell mitogen

responses (Field, McBurney, Massimino, Hayek, and Sunvold, 1999), while other

studies have reported on the ability of insoluble fibres to protect the colonic mucus

barrier (Brownlee, 2011; Brownlee, Dettmar, Strugala and Pearson, 2006; Meguro,

Hase, Tokimitsu and Sakata, 2001). Although there are numerous reports on the

prebiotic and probiotic effects of dietary fibres, especially concerning infant nutrition,

the research surrounding the health benefits of insoluble fibres is ongoing. Similarly,

researchers are focusing on the events that take place in the upper GIT, in hopes of

better understanding and explaining the beneficial effects of soluble fibre

consumption.

Fermentability is a physicochemical property associated with insoluble dietary

fibres and some soluble fibres, such as guar gum and xanthan gum. In addition to this,

soluble fibres also have the ability to induce viscosity in the upper GIT, which is

linked to a variety of physiological responses. The degree of thickening is dependent

on the inherent properties of different polysaccharides, such as molecular weight

(MW), degree of polymerisation, chemical composition and overall structure, as well

as concentration in solution. For example, a study was done that investigated the

34

viscosity of solutions prepared using guar gum flours with different MW. The

researchers found that at 40 g/kg solution, viscosities ranged from 0 cP for the low

MW, 1,600 cP for the medium MW and 8,200 cP for the high MW flours (Roberts,

Smith, Low, Ellis, Morris, and Sambrook, 1989). This again highlights that not all

fibres are alike, and although soluble fibres may act similarly when exposed to the

environment inside the GIT, some may be more effective than others in mediating a

physiological response.

The confounding evidence of the physiological effects of soluble fibre

consumption has allowed the FDA to place a health claim on food products containing

soluble fibre, that reads “soluble fibre from foods such as (name of soluble fibre

source and/or name of food product), as part of a diet low in saturated fat and

cholesterol, may reduce the risk of heart disease” (FDA, 2013). Currently, Health

Canada has permitted two claims that are fibre-specific, “oat fibre helps reduce/lower

cholesterol, (which is) a risk factor for heart disease”, and ”psyllium fibre helps

reduce/lower cholesterol” (Health Canada, 2011). A recent meta-analysis on 22 cohort

studies concluded that foods rich in fibre, specifically from cereal or vegetable

sources, lead to substantial reductions in the risk of developing cardiovascular disease

(Threapleton, Greenwood, Evans, Cleghorn, Nykjaer, Woodhead, Cade, Gale, and

Burley, 2013). There is growing evidence to support the position that consumption of

foods rich in dietary fibre are useful in combating other major health problems and

their pre-determining risk factors, such as high cholesterol levels, obesity, certain

types of cancer, and T2D. The results of a human intervention trial demonstrated that

the addition of oat beta-glucan has a substantial effect on lowering plasma cholesterol.

35

The researchers concluded that high MW, and therefore higher viscosity, was

responsible for this observation, indicated by a 50% reduction in the observed effects

when subjects were fed lower MW beta glucans (Wolever, Tosh, Gibbs, Brand-Miller,

Duncan, Hart, Lamarche, Thomson, Duss, and Wood, 2010). Several in vitro studies

have been undertaken in attempt to explain the effects of soluble fibre on cholesterol

reductions. In particular, researchers focus on the interactions between soluble fibre

and bile acids/bile salts, indicating that there is a binding effect involved (Kahlon,

Chiu, and Chapman, 2009). Moreover, there are reports that the presence of viscous

fibres in the small intestine leads to excess excretion of bile salts in the faeces. This

depletes the storage of bile acids in the liver and promotes catabolism of cholesterol in

the hepatocyte in order to replenish the bile acid pool (Marlett and Fischer, 2003;

Ellegard and Andersson, 2007; Ellegård, Andersson, Normén, and Andersson, 2007).

Gunness and Gidley (2010) further explored this through the use of a dialysis

membrane in order to determine whether a viscous barrier is formed thereby

preventing mixed micelles from being absorbed across the basolateral membrane, as

illustrated in Figure 2-6.

36

Figure 2-6. Diagram of the potential effect of viscous soluble dietary fibre (SDF) on

reducing micelle absorption (Adapted from Gunness and Gidley, 2010).

As a follow up to their study, they showed that inclusion of soluble dietary fibre,

namely barley beta-glucan and wheat arabinoxylan, led to delayed transport of bile

salts across a semi-permeable membrane, which they believe was attributed to by

increasing concentrations of fibre and therefore increasing viscosity (Gunness,

Flanagan, Shelat, Gilbert, and Gidley, 2012).

In addition to the effects on cholesterol lowering, fortification of foods with

soluble fibres has been shown to be an effective strategy in reducing both blood

pressure (Streppel, Arends, van ‘t Veer, Grobbee, and Geleinjse, 2005) and weight

loss to counteract the increase incidence in obesity (Brownlee, 2011). Recently, a

37

clinical trial was carried out where researchers fortified orange juice with high-

methoxyl, low-methoxyl, and low-methoxyl amidated pectin. They observed that

following simulated digestion (Fabek, 2011) there were substantial reductions in

viscosity of all solutions. Subsequently, this affected the satiety response in human

subjects wherein the low-methoxyl pectin was the only treatment that resulted in

elevated feelings of satiety (Logan, Wright, and Goff, 2015). Investigators of a seven

day randomised crossover trial of 68 participants indicated that beverages fortified

with 1.5 g sodium alginate led to significant reductions in daily energy intake in

comparison to the control. They attributed their findings to alginate-induced gelation

inside the stomach, which therefore reduced gastric emptying rates (Paxman,

Richardson, Dettmar, and Corfe, 2008). On the other hand, researchers using echo-

planar magnetic resonance imaging have demonstrated that nutrient composition is

more important in slowing gastric emptying than viscosity. In their four-way

crossover design, which consisted of low-viscosity nutrient control, high-viscosity

nutrient control, low-viscosity nutrient and high-viscosity nutrient meals, they found

that not only was gastric emptying inversely related to viscosity, but the presence of

nutrients led to the greatest reductions in subsequent rates that were observed

(Marciani, Gowland, Spiller, Manoj, Moore, Young et al., 2001). Moreover, the

researchers, in conjunction with their earlier work (Marciani, et al., 2000), observed

that intragastric secretions intensified when there was greater viscosity and nutrient

composition in the body of the stomach. They hypothesize that the stomach works

harder to dilute the viscosity and that greater distension of the stomach may also be a

precursor to the rate of gastric secretions. Although the effect of fibre on satiety

38

remains a contentious topic, researchers have demonstrated that soluble fibres are

more effective in reducing appetite between meals in comparison to other dietary

interventions, including high-fat diets (Slavin and Green, 2007; Halton and Hu, 2004).

Dietary fibre intake has also been associated with reductions in postprandial glycemia

(Jenkins et al., 1987), making it a staple ingredient in the management of T2D.

Results of both animal and human studies conducted over the past couple of

decades have highlighted that diets high in soluble fibre improve overall glucose and

glycated protein levels in the blood and lead to increased insulin sensitivity, which are

all biomarkers for glycemic control (Kendall et al., 2010; Brennan, 2005; Schulze and

Hu, 2005; Behall, Scholfield, and Hallfrisch, 2005; Brennan and Cleary, 2005; Judd

and Ellis, 2005; Gallaher and Schaubert, 1990). These findings are further supported

through a recent meta-analysis that looked at the effects of high fibre diets vs. low

fibre diets on biomarkers for glycemia (Anderson et al., 2004). Despite evidence

suggesting that soluble fibre consumption reduces postprandial plasma glucose levels

by way of the viscosity impacting rates of gastric emptying, many studies have shown

that the effects inside the small intestine are more prominent. Edwards et al., (1987)

measured gastric emptying in human volunteers by monitoring changes in

radioactivity of a sulfur colloid that was consumed along with 50 g glucose test drinks

with and without added fibre (2.5 g of 1% locust bean gum, xanthan gum, or a

combination of both). The results indicated that in the control group 28% of the drink

had emptied into the stomach after 30 min and this value was reduced to 11% for the

locust bean gum/xanthan gum mixture. Interestingly, no difference was observed

among individual gums and the control, yet all treatments were able to improve the

39

glucose response in comparison to the control. The authors thereby concluded that

effects on gastric emptying alone may not be responsible for the modulating effects of

soluble fibres on the glycemic response in vivo. An earlier study resulted in a similar

finding, wherein different doses of guar gum (2.5 g or 14.5 g) were added to glucose

test drinks. Gastric emptying was measured in a similar fashion, using changes in

radioactivity, followed by 60 min postprandial glucose measurements. The authors

concluded that neither dose of guar gum led to changes in gastric emptying, yet both

doses were able to significantly reduce the levels of plasma glucose that were

measured in comparison to the control (Jarjis, Blackburn, Redfern, and Read, 1984).

More commonly, in vitro experiments are employed to analyse fibre

functionality in the GIT, due to the aforementioned costs and ethical considerations of

in vivo work. As was shown in Section 2.1.2, starch digestion and carbohydrate

metabolism are complex processes, making it difficult for researchers to determine

how viscosity is able to modulate the glycaemic response using simulated models.

Nonetheless, a commonly employed technique is using dialysis tubing to mimic

glucose mobility across a semi-permeable membrane. Sasaki and Kohyama (2011)

demonstrated that formulation of rice starch gels with non-starch polysaccharides

(NSPs), agar, xanthan gum and konjac glucomannan, leads to reduced starch

digestibility. The authors demonstrated that not only did gel rigidity increase but they

hypothesize there is an interaction between the starch and the gums and that the

addition of NSP’s leads to greater viscosity, which altogether may contribute to the

reductions in digestibility that they observed. In a more recent study, researchers

explored the effects that adding xanthan gum, guar gum, konjac glucomannan, and

40

pectin would have on glucose diffusion. Their results indicated that all gums were

able to promote viscosity in solution. Moreover, an inverse relationship between

viscosity and glucose diffusivity was established, whereby greater concentrations of

NSPs led to more pronounced suppressive effects on diffusion, with xanthan gum

being the most effective at the same concentration (Sasaki and Kohyama, 2012).

Clinical evidence of soluble fibre functionality exists, demonstrating that formulation

of foods and drinks with beta-glucan leads to reductions in glycemic responses in

humans (Juvonen, Purhonen, Salmenkallio-Marttila, Lähteenmäki, Laaksonen,

Herzig, Uusitupa, Poutanen, and Karhunen, 2009; Panahi, Ezatagha, Temelli,

Vasanthan, and Vuksan, 2007; Brennan and Cleary, 2005; Jenkins et al., 2002).

Subsequently, a study was carried out that examined whether solutions and gels

formulated with low-MW and high-MW oat beta-glucan have any effects on lowering

glucose mobility, in vitro, and reducing glycemic responses, in vivo. The in vitro

results showed that all beta-glucan gels released significantly less glucose in

comparison to the beta-glucan solutions. The researchers attributed their findings to

the gels physically trapping the glucose within the food matrix, therefore leading to

reductions in the amount of glucose permeating through the dialysis membrane.

Interestingly, the in vivo results did not demonstrate the same trend. Fifteen

participants received test meals, which were either liquid solutions or viscoelastic

gels, each containing either low-MW or high-MW oat beta-glucan. The high-MW

solutions elicited the lowest glycemic response; however, the gels did not differ from

the control. The authors explained their findings by looking at the rheological

properties of each treatment. Analyses of the storage modulus (G’) and loss modulus

41

(G”) and frequency dependence tests showed that the oat beta-glucan drinks may be

abundant in polymer chain entanglements. On the other hand, the gels were more

abundant in hydrogen bonds, which could have been disrupted through chewing and

mechanical mixing, and the openness of the gel networks resulted in an inability to

attenuate the glucose response. A measureable viscous component (G”) seems to be

vital to the functionality of oat beta-glucan and polymer chain entanglements appear

to be necessary for reducing glucose diffusion. In addition, the researchers of this

study speculate that physiological functionality of oat beta-glucan is not only due to

viscosity being able to lower diffusion of molecules, but may also be a result of

impairment of mixing with digestive enzymes and reduced gastric emptying (Kwong,

Wolever, Brummer, and Tosh, 2013). Other studies suggest that soluble dietary fibre

acts like a sponge, forming an entangled network with the food through enhanced

water absorption, which thereby acts to slow the rate of starch digestion by impairing

both enzyme transport to its substrate (Gidley, 2013) and sugar absorption across the

intestinal epithelium (Singh, Dartois, and Kaur, 2010; Wursch and Pi-Sunyer, 1997).

To further complicate our understanding concerning the functionality of soluble fibre,

it has also been suggested that an enhancement of the ileal brake may be involved.

This would act to reduce gastric emptying and acid secretion, small intestinal transit,

pancreatic enzyme secretion and bile acid secretion, which altogether may play a role

in some of the proposed health benefits that arise from fibre consumption (Kristensen

and Jensen, 2011; Maljaars, Peters, and Masclee, 2007). Nonetheless, it is evidenced

in literature that using in vitro methods is a vital tool in allowing researchers to assess

42

the physiological functionality of soluble dietary fibres in order to gain a clearer

understanding of structure-function relationships.

Despite emerging evidence that soluble fibre intake is related to reduced

postprandial glycemic levels, functional dissimilarities are a result of structural

diversity present among different fibres. Viscosity results from physical

entanglements of polysaccharide chains (Dikeman and Fahey, 2006). Therefore,

chemical composition, overall structure (linear versus branched), MW, as was shown

with oat beta-glucan, and degree of polymerisation (DP) all play a role in the ability of

fibres to induce viscosity in solution. As an extension, these structural differences

have a bearing on their ability to exert physiological functionality when consumed as

a food ingredient.

2.3. Soluble fibres employed in the study

2.3.1. Guar Gum

Guar gum belongs to the family of reserve carbohydrates known as

galactomannans, which are storage polysaccharides found in the cell wall of various

albuminous and endospermic seeds. One of the most widely used seeds comes from

the guar plant (Cyamopsis tetragonoloba). The plant is bush-like growing up to 90 cm

tall and is drought resistant, allowing it to thrive in arid and semi-arid zones. It has

been grown for centuries in India and Pakistan where it is widely used as a food

source for humans and animals. However, due to the increased demand of guar

production, programmes have been established in other parts of the world, including

Australia, Colombia, Argentina and some parts of North America, such as Texas,

43

Oklahoma and Arizona, due to the favourable environmental and economic conditions

present there. The guar seeds are composed of hull (20 – 22% wt.), germ (43 – 44%

wt.) and endosperm (34 – 36% wt.). Although the germ layer is nutritionally-rich,

consisting of more than 50% wt. protein, it is the endosperm that contains the majority

of the gum (Wielinga and Maehall, 2000). Guar seeds are split into endosperm halves,

which are further processed to commercial powder, with varying viscosities and

particle sizes, using a variety of techniques, including milling and screening, as shown

in Figure 2-7.

Figure 2-7. Flow diagram illustrating guar gum production (Adapted from Wielinga,

2010).

Subsequently, the inherent structure of guar gum allows it to be used as a food

ingredient for a variety of different purposes.

44

Guar gum consists of a linear 1, 4 linked-β-D-mannan backbone, as do all

galactomannans; however the differentiating factor among the galactomannans is the

degree of substitution along the mannose backbone. Guar gum, unlike locust bean

gum and tara gum, has a 1, 6 α-linked galactose side chain at every other mannose

unit (Kulicke, Eidam, Kath, Kix, and Kull, 1996), as shown in Figure 2-8. However,

as many as 5 unsubstituted regions may be found along the backbone (Wielinga and

Maehall, 2000).

Figure 2-8. Chemical structure of the galactomannan backbone (Adapted from

Wielinga, 2010).

The molecular weight of guar gum ranges between 50,000 to 8,000,000 Da

(Kawamura, 2008) and DP is 300 – 50,000 hexoses. With increasing concentrations in

solution there is an increase in viscosity and rheological flow behaviour displays

solutions that behave as non-Newtonian, pseudoplastic fluids. One of the main

functional properties of guar gum is linked to the ability to promote viscosity in

45

solution, due to the interactions between polymer chains. However, this also acts to

limit potential uses for guar gum in foods. Typically, the food industry incorporates

guar gum at a concentration of ≤1% to facilitate gelling, emulsifying, thickening and

prevent creaming, syneresis and retrogradation of starch. It is used in foods such as ice

creams and other dairy products such as puddings, soft drinks, baby foods and pet

foods (Wielinga, 2010). Because measurably high viscosity, achieved at

concentrations greater than 1%, affects textural and organoleptic properties of foods,

partially hydrolysed guar gum is often used to circumvent such undesirable

characteristics (Roberts, 2011). However, this may impact the multi-functional

behaviour of guar gum as it has been shown to also promote physiological

functionality upon consumption through increasing digesta viscosity.

An earlier study was conducted to analyze the effect that three viscous

hydrocolloids, guar gum, pectin, and carboxymethylcellulose, would have on

postprandial glycemia in humans. The results indicated that despite comparable initial

viscosities, guar gum showed greater resistance to in vitro digestion and also led to

more pronounced reductions in plasma glucose levels, as measured in vivo (Brenelli,

Campos, and Saad, 1997). Gularte and Rossell (2011) reported that blending of guar

gum with potato starch led to pronounced reductions in the amount of starch that was

hydrolysed over a 3 h simulated digestion period. They attribute their findings to the

potential physical barrier that could act to hinder α-amylase access to the substrate.

Similar results were obtained from a study that examined predicted glycemic index

(PGI) values of cereal foods through soluble fibre supplementation. The findings

indicate that adding guar gum at levels of 2.5% and 5% reduced the PGI values by 4

46

and 13%, respectively, in breads and those values increased to 37% and 43% in pasta

(Brennan, 2008). Similarly, researchers have shown that guar gum supplementation

also leads to reduced starch hydrolysis in durum wheat through enhanced viscosity of

the flour pastes (Brennan and Tudorica, 2008) and supplementation in breakfast

cereals reduced the amount of RDS while increasing the quantity of SDS (Brennan,

Monro, and Brennan, 2008). Results of animal studies also suggest that guar gum

supplementation and the ensuing increase in digesta viscosity allow for improved

glycemic control (Owusu-Asiedu, Patience, Laarveld, Van Kessel, Simmins, and

Zijlstra, 2006; Cameron-Smith, Collier, and O’Dea, 1994) and reduced levels of

glycated hemoglobin (Gallaher and Schaubert, 1990). In addition to influences on

plasma glucose concentrations, there is clinical evidence supporting the beneficial

effect of guar gum on serum cholesterol levels (Yamatoya, Kuwano, and Suzuki,

1997), measured lipemia (Redard, Davis, and Schneeman, 1990), reducing diarrhea

and constipation (Homann, Kenen, Fussenich, Senkal, and Zuntobel, 1994; Takahashi,

Wako, Okubo, Ishihara, Yamanaka, and Yamamoto, 1994) and promoting propionate-

rich fermentation in humans. The multi-faceted nature of guar gum makes it a highly

researched food ingredient, due to its ability to impart both technological as well as

physiological functionality in food systems.

2.3.2. Xanthan Gum

The name ‘xanthan’ comes from the bacterium from which it is produced,

Xanthomonas campestris. The bacteria may be found on the leaves of Brassica

vegetables, such as cabbage, alfalfa, and beans, and are used in the commercial

production of xanthan gum through an aerobic fermentation process, as shown in

47

Figure 2-9. In short, the microbial strain is selected and then preserved to allow for

long-term storage, the bacteria are cultured in the presence of glucose, nitrogen and

trace elements such as potassium, iron, and calcium. Inoculum build-up is further

carried out in several stages that provide the seed for the final fermentation. In the

end, the broth consists of xanthan, bacterial cells and chemicals (Garcia-Ochoa,

Santos, Casas, and Gomez, 2000). Subsequently, the cells are removed by filtration or

centrifugation and the remaining bacteria are killed by pasteurisation of the broth at

which point the gum fraction is further extracted by precipitation using isopropyl

alcohol (Sworn and Monstanto, 2000). After precipitation, the product is dehydrated

and dried and subsequently milled and packaged to allow for distribution. Similar to

guar gum, it is the structure of the polysaccharide that dictates its use in foods.

Figure 2-9. Flow diagram illustrating xanthan gum production (Adapted from Garcia-

Ochoa et al., 2000).

48

Xanthan gum is a heteropolysaccharide consisting of a linear 1,4 – linked β-D-

glucose backbone with a trisaccharide side chain, containing D-glucuronic acid units,

linked at the O-3 position of every other glucose unit, as illustrated in Figure 2-10.

Figure 2-10. Structure of xanthan gum (Adapted from Garcia-Ochoa et al., 2000).

The molecular weight ranges from 2x106 Da to 2x10

7 Da, depending on the

fermentation conditions used in the production process (Garcia-Ochoa et al., 2000). It

is an anionic polysaccharide due to the presence of acetic and pyruvic acids. The

trisaccharide branches align with the main backbone inducing a stiff single, double, or

triple helix (Cairns, Miles, and Morris, 1986; Rinaudo and Milas, 1982; Milas and

Rinaudo, 1979) and this helical formation has been shown to provide a protective

barrier towards pH extremes and accessibility to hydrolytic enzymes (Sworn and

Monsanto, 2000). Moreover, the ordered helical conformation allows for synergistic

interactions with other gums, such as galactomannans, allowing for gel formation and

greater solution viscosities to be attained (Renou, Petibon, Malhiac, and Grisel, 2013).

Xanthan gum solutions have non-Newtonian rheology, wherein the apparent viscosity

49

decreases with increasing shear rates. Despite both shear-thinning and recovery being

instantaneous (Kang and Pettit, 1993), an initial yield stress is required in order for

xanthan gum solutions to flow, providing storage and transportation stability for

emulsions prepared with xanthan gum. These unique properties and characteristics

make xanthan gum a popular ingredient used by the food industry, where it may act as

a thickener, dispersant, and stabilizer in dairy products, baked goods, frozen foods,

beverages and salad dressings. In addition to this, results of several studies have

suggested that xanthan gum is able to promote physiological functionality by acting as

a source of dietary fibre.

A recent study explored the effects that different hydrocolloids, including

xanthan gum, have on starch digestibility. The results demonstrated that although

some of the hydrocolloids increase the RDS fraction of the starch, xanthan gum was

able to decrease C∞, which in this study was used as the equilibrium concentration of

hydrolyzed starch, therefore indicating that adding xanthan gum decreased the

maximum hydrolysis of starch (Gularte et al., 2011). In a similar study, researchers

explored how adding non-starch polysaccharides would affect in vitro digestibility of

starch suspensions. Their results indicated that xanthan gum had the most pronounced

effect on suppressing starch digestion in comparison to the other gums used in the

study. They attributed their findings to not only the increase in digesta viscosity but

also a potential interaction that could exist between xanthan and starch, which would

encapsulate the starch granule creating a barrier to hydrolytic enzymes. However, they

indicated that this needs to be explored further (Sasaki and Kohyama, 2012). The

effect on starch digestion was also observed in an earlier study that showed reduced

50

hydrolysis of starch through supplementation of xanthan gum in pasta (Brennan et al.,

2008). Moreover, consumption of viscous gums such as xanthan has been associated

with reductions in plasma cholesterol (Dikeman et al., 2006) and increases in satiety

(Sworn, 2010). Although widely used in the food industry, more remains to be

elucidated regarding the physiological functionality of xanthan gum.

2.3.3. Flaxseed Gum

Flaxseed (Linum usitatissimum L.) was first introduced in North America as a

fibre crop, however over time it has become of great importance as a source of oil to

consumers and food manufacturers (Cunnane and Thompson, 1995). Canadian

production accounts for more than 17% of the global production of flaxseed (Statistics

Canada, 2012) and is considered to be the richest plant source of α-linolenic acid in

the North American diet (Morris, 2001). Although flax contains nearly 45% oil, which

is found in the endosperm, it is also comprised of soluble flaxseed gum, which can be

isolated from mucilage found inside the hull (Cui and Mazza, 1996). In addition, the

kernel of flaxseed has been shown to contain nearly 20% dietary fibre, the structure of

which was recently elucidated through the use of high performance anion exchange

chromatography. It is mainly comprised of glucose, xylose, galactose, and arabinose

and the molecular weight was shown to range from 486 kDa to 1660 kDa (Ding, Cui,

Goff, Wang, Chen, and Han, 2014).

Soluble (mucilage) flaxseed gum is reported to be comprised of neutral and

acidic fractions, and ion-exchange chromatography reveals a high molecular weight

arabinoxylan for the neutral fraction and a low molecular weight pectin-like

51

rhamnogalacturonan for the acidic fraction (Cui et al., 1994). Their structures dictate

their rheological properties in solution. It is reported that the different fractions behave

as Newtonian fluids at low concentrations and that pseudoplasticity is not observed

until the concentration exceeds 0.2%. However, for acidic fractions Newtonian

behaviour has been reported for concentrations of up to 2%, and has been observed to

be substantially less viscous than the neutral fraction (Mazza and Biliaderis, 1989).

Viscosities reported for flaxseed mucilage gum (Qian, Cui, Wu, and Goff, 2012) and

flaxseed kernel gum (Ding et al., 2014) are substantially lower than other

commercially available gums, such as guar gum and xanthan gum (Fabek, 2011).

These rheological properties allow for functionality in food systems where they are

added as thickeners, emulsifiers, stabilisers, and fat replacers without having adverse

effects on organoleptic properties.

Furthermore, studies have shown that inherently low GI values allow flaxseed

to be used in value-added foods such as salad dressings, pasta, muffins and dairy

products (Hall, Tulbek, and Xu, 2006; Lee, Manthey, and Hall, 2003). Soluble

flaxseed gum has also been associated with increased bile-acid binding (Fodje, Chang,

and Leterme, 2009) and has been shown to reduce plasma glucose and cholesterol

concentrations in human trials (Thakur, Mitra, Pal, and Rousseau, 2009). Moreover,

animal studies have demonstrated that flaxseed gum exerts a prebiotic effect through

the production of acetate and propionate (Fodje et al., 2009) and stimulation of the

growth of lactobacilli (Alzueta, Rodriguez, Cutuli, Rebole, Ortiz, Centeno, and

Trevino, 2003). Despite unremitting research efforts to elucidate the structure of

52

soluble flaxseed gum, more remains to be discovered of its potential physiological

effects and subsequent behaviour along the GIT.

2.3.4. Soy soluble polysaccharide

Soybean soluble polysaccharide (SSPS) is a water-soluble by-product that is

obtained during production of tofu (soybean curd), soy milk, and soy protein isolate

(Chivero, Gohtani, Ikeda, and Nakamura, 2014). Okara is an insoluble residue that is

produced from protein extraction and it is this insoluble fraction that leads to the

production of SSPS. In short, the material is autoclaved in weak acidic conditions,

followed by processes of refining, pasteurising and spray-drying to produce a white

powder. SSPS powders have been in the market since the early 1990’s when Fuji Oil

marketed the product as SOYAFIBE-S. Its structure allows for comparably low

viscosity and stability in aqueous solutions (Maeda, 2000).

SSPS consists of a rhamnogalacturonan and homogalacturonan backbone with

branched β-1,4-galactan and α-1,3- or α-1,5 side chains, along with monosaccharide

constituents, fructose, xylose, and glucose embedded along the main backbone

(Nakamura, Furuta, Maeda, Takao, and Nagamatsu, 2002), as illustrated in Figure 2-

11.

53

Figure 2-11. Structure of SSPS (adapted from Nakamura, et al., 2002).

Subsequently, HPLC analysis reveals a different MW for each fraction, ranging from

550,000 Da for the major component, to 25,000 and 5,000 Da for the minor

constituents. It has a spherical structure that assumes the conformation of flexible

random coils (Chivero et al., 2014) with dimensions smaller than 80 nm (Nakamura,

Fujii, Tobe, Adachi, and Hirotsuka, 2012) and contains 66% dietary fibre, 9% protein

and 8.6% ash with a self-affirmed status of GRAS in North America.

The unique structural properties of SSPS allow it to be used in flavour

emulsions, due to high water solubility, and it also has excellent pH stability where it

is extensively used as a stabiliser in acidic beverages, such as in yogurts. Although

pectin is a commonly used stabiliser, SSPS is able to prevent protein particles from

54

coagulating and precipitating out of solution without raising the viscosity of the

product thereby maintaining desirable product characteristics. Moreover, it is able to

form strong interfacial films (Furuta and Maeda, 1999), prevent oxidation of oils, and

has excellent thermal stability and emulsifying properties, permitting it to be used in

foods such as ice cream, dairy products, baked goods and dressings (Nakamura,

Yoshida, Maeda, and Corredig, 2006). Furthermore, its reported high fibre content is

spurring research interests in exploring potential physiological functionality

surrounding SSPS consumption.

Studies have shown that consumption of soy fibre leads to a stool bulking and

laxation effect due to colonic fermentation (Schneeman, 1998; Wrick, Robertson, Van

Soest, Lewis, Rivers, Roe, and Hackler, 1983). A much earlier diet-controlled study

demonstrated that supplementing 25 g/d with SSPS led to significant improvements in

glucose tolerance as well as increased fecal wet weight and water content. Moreover,

supplementation with SSPS led to improved nutrient balance through increasing

amounts of protein, calcium, iron and other minerals in the diet (Tsai, Mott, Owen,

Bennick, Lo, and Steinke, 1983). In a separate clinical trial, SSPS consumption led to

significant reductions in the duration of liquid stool excretion in infants suffering from

diarrhea (Brown, Perez, Peerson, Fadel, Brunsgaard, Ostrom, and MacLean, 1993).

Although SSPS consists primarily of dietary fibre, very little is known about the

functionality of SSPS in the human GIT. A recent study showed that addition of 6%

SSPS and 0.7% flaxseed gum, allowing for equal solution viscosities, to 50 g glucose

solutions had no effect on measured postprandial glucose and insulin levels in humans

(Au, Goff, Kisch, Coulson, and Wright, 2013).. However, differences in glucose AUC

55

were reported when the fibres were added to dairy beverages and puddings, with no

differences observed between the dairy product types. They concluded that product

viscosity was responsible for mediating a physiological response, not the

concentration in solution. Moreover, they suggest that despite there being a

relationship between viscosity and glucose attenuations, the fibre polymers are subject

to structural changes when passing through the GIT. Subsequently, the viscosity

measured in solution may not be indicative of the physiological viscosity in the small

intestine (Au et al., 2013). This raises a critical question surrounding the functionality

of soluble fibres, whether they may are extensively studied, such as guar gum and

xanthan gum, or understood to a lesser degree such as SSPS and flaxseed gum.

2.3.5. Summary Comment

Viscosity is a physicochemical property that is intrinsic to all soluble fibres

(Dikeman, Murphy, and Fahey Jr., 2006). There is extensive information present in

the literature that supports the claim that viscous soluble fibres are able to modulate

the glycemic response in humans. However, the GIT is exceedingly complex and for

that reason the mechanistic action remains to be elucidated. In order to fully

understand how dietary fibres are able to promote physiological functionality, it is

vital to look at and understand how food is metabolised by employing a variety of

analytical techniques, including the aforementioned in vitro digestion method.

56

3. The effect of in vitro digestive processes on the viscosity of

dietary fibres and their influence on glucose diffusion

Hrvoje Fabeka, Stefanie Messerschmidt

a, Valentin Brulport

a, H. Douglas Goff

a,*

a Department of Food Science University of Guelph, Guelph, Ontario N1G 2W1,

Canada

Parts of this chapter were published in 2014: Food Hydrocolloids, 35, 718-726.

Abstract

The consumption of soluble dietary fibres is associated with a variety of physiological

responses, one of which is their ability to modulate postprandial glycemic response

due to increasing digesta viscosity. The effects that the different digestive processes in

the stomach and small intestine have on the thickening ability of six soluble fibres

were investigated in this study. Moreover, the ability of these fibres to resist loss of

viscosity following two-stage in vitro digestion, simulating gastric and small intestinal

phases, was analyzed. In a second set of experiments, protein and starch were

included to simulate a food model, which allowed for glucose release. The second

stage was carried out inside a dialysis system. The primary determinant of starch

digestibility was the concentration of glucose in the dialysate over a 3 h period.

Irrespective of the concentration level or simulated condition, xanthan gum (XG)

retained viscosity more than all other fibre types (p<0.05). XG reduced glucose

concentrations in the dialysate in comparison to the control and the others fibres

employed in this study (p<0.05). The ability of XG to attenuate glucose mobility

means that it may be a target for future studies, both in vitro and in vivo, which seek to

57

analyze food ingredients as aids in effectively mediating the postprandial glycaemic

response in humans.

3.1.Introduction

In recent years, dietary fibre has received a great deal of attention from

researchers, the food industry, and consumers, due to the health benefits that are

associated with the consumption of fibre-rich foods (Brighenti, Pellegrini, Casiraghi,

& Testolin, 1995). Beneficial effects include lowering blood lipid levels, specifically

triglycerides and low-density lipoprotein cholesterol, reduction in risk of

cardiovascular disease, increases in satiety and consequent decreases in obesity trends,

enhanced gastrointestinal immunity and overall colonic health, and decreases in blood

glucose levels (Jenkins, Leeds, Gassull, Cochet, & Alberti, 1977; Jenkins, Kendall,

Axelsen, Augustin, & Vuksan, 2000; Gunness & Gidley, 2010; Brownlee, 2011).

However, despite the numerous studies and growing consumer awareness surrounding

the health benefits of dietary fibre consumption, average intake levels still fall far

below recommended levels (Health Canada, 2010). Selecting fibre for food

fortification may therefore be a fundamental strategy for the food industry.

Fibre molecules possess distinctive physicochemical characteristics that dictate

their uses in foods, and are categorized as being soluble or insoluble (Dikeman &

Fahey, 2006). The inherent solubility is what determines their fate as they pass

through the gastrointestinal tract (GIT). For insoluble fibres (e.g., cellulose, wheat

bran), it is their ability to ferment that allows them to act as functional foods. On the

other hand, soluble fibres (e.g., gums, pectins) rely on their ability to thicken into

58

swollen hydrated networks and their ensuing viscosity that dictates their potential to

exert biological effects along the GIT, specifically through the stomach and small

intestine (Schneeman, 2001; Brownlee, 2011). Dikeman, Murphy, & Fahey Jr. (2006)

showed that soluble fibres, such as guar gum and psyllium, promote relatively high

viscosities in comparison to insoluble fibres such as cellulose, rice bran, and wheat

bran. Consequently, they proposed that because of this relationship, soluble fibres

may have an effect on postprandial blood glucose concentrations. Accordingly,

hydrocolloids can act as soluble dietary fibre. Their uses may extend beyond the

technological functionalities they currently promote in foods, (e.g., gelation, product

consistency, colloidal stability, texture, and viscosity) if they have the potential to

deliver a health benefit to the consumer without having negative impacts on

organoleptic properties.

There are several proposed mechanisms by which viscosity may elicit

physiological responses such as attenuation in postprandial glycaemia. These include

increases in luminal bulk, changes in neurohumoral mediation, and impairing nutrient

diffusion across the unstirred boundary layer of the mucosal membrane (Chawla &

Patil, 2010). Starch digestion and subsequent glucose absorption occur mainly in the

small intestine (Hasjim, Lavau, Gidley, & Gilbert, 2010). In the lumen, starch is

hydrolyzed by α-amylase, secreted from the pancreas. The resulting oligosaccharides

and maltose are then further hydrolyzed by brush border enzymes (Tharakan, Norton,

Fryer & Bakalis, 2010), leading to glucose transport through the basolateral

membrane into the blood (Guyton & Hall, 2010). Therefore, an increase in digesta

viscosity that arises from soluble fibre consumption may influence one or more events

59

that occur during starch metabolism and glucose uptake. However, the stability of

both the polysaccharide networks and the various non-covalent bonds that stabilize the

junction zones of the polymers are susceptible to the conditions of the GIT (e.g.,

secretion of fluids, enzyme hydrolysis, pH extremes, and the presence of salts).

Eastwood & Morris (1992) mentioned specifically how charged polysaccharides are

precariously sensitive to pH, and that polysaccharide networks in general are

influenced by other factors present along the GIT, including bile salt concentrations.

Bobboi & Stephens (1996) demonstrated that addition of guar gum to glucose

containing solutions led to increases in viscosity. However, they observed reductions

in apparent viscosities at 150 s-1

after acidification. Evidently, physiological

conditions, such as pH levels, will affect the viscosity of soluble fibres and as an

extension may alter their ability to modulate nutrient absorption in the foregut.

Moreover, the ability of polysaccharide solutions to enhance viscosity is a direct result

of the inherent ability of polymer chains to overlap with each other creating a greater

number of junction zones. The degree of thickening is therefore reliant on the

hydrodynamic volume, or intrinsic viscosity, and the number of coils present in

solution, and is not universal among all soluble fibres (Eastwood & Morris, 1992).

The viscosity of branched polysaccharides is lower in comparison to long linear

polymers that are able to promote greater entanglements when placed in solution.

Therefore, in order to select appropriate hydrocolloids for food fortification to

enhance their dietary fibre levels, it is essential to not only have an understanding of

the viscosity characteristics of an array of polysaccharides and biopolymers, but more

60

importantly recognize their ability to withstand postprandial changes that lead to

reductions in digesta viscosity.

The objectives of this study were to evaluate the viscosity profiles, across a range of

shear rates, for four structurally-different hydrocolloids at various concentrations both

in solution and during a 2-stage in vitro digestion. A second objective was to

determine the impact that fibres have on lowering the diffusion of glucose using a

dialysis system.

3.2. Materials and methods

3.2.1. Materials

Four hydrocolloids, including guar gum, (Danisco Canada Inc., Toronto, ON,

Canada), xanthan gum (Sigma Chemical CO., St. Louis, MO, USA), soluble flaxseed

gum extracted from flaxseed hulls (refer to Section 3.2), and DA-100 variety soy

soluble polysaccharides (Fuji Oil Co. Ltd., Osaka, Japan), were tested. Waxy corn

starch, ULTRA-SPERSE® A, was purchased from National Starch (Bridgewater, NJ,

USA). Sodium caseinate (ALANATE® 180) was purchased from NZMP Inc.

(Lemoyne, PA, USA). Study-specific enzymes and chemicals were purchased from

the following distributors: simulated gastric fluid (SGF) from Ricca Chemical

Company (Arlington, TX, USA), purified pepsin, sodium phosphate monobasic

monohydrate and sodium phosphate dibasic anhydrous from Fisher Scientific (Fair

Lawn, NJ, USA), pancreatin from MP Biomedicals (Solon, OH, USA),

amyloglucosidase (Aspergillius niger) from Sigma-Aldrich (St. Louis, MO, USA),

hydrochloric acid 2N solution from Fisher Scientific (Nepean, ON, Canada), bile salts

from Fisher Science Education (Hanover Park, IL, USA).

61

3.2.2. Soluble flaxseed gum extraction

Omega-3 flaxseed hulls (Natunola Health Biosciences Inc., Winchester, ON,

Canada) and deionized water were used for flaxseed gum extraction. Flaxseed hulls

(3.0 kg) were allowed to soak in 39 L of deionized water. The mixture was stirred for

18 h at 300 rpm and 20oC, using an electrical stirrer (Model 4100, Burlington Pump

Inc., Burlington, ON, Canada) to extract the soluble fibre. The extract was filtered

using 4 layers of cheese cloth to separate the desired mucilage from the hulls. The

mucilage was transferred into 500 mL centrifuge tubes whereby the flaxseed gum was

separated from any remaining insoluble fractions by using a Beckman JA-10 Rotor in

a JA-21 centrifuge (Beckman Coulter Inc., Palo Alto, CA, USA) at 9000 rpm, 20oC

for 20 min. The remaining soluble fractions were then poured onto trays, which were

subsequently placed in a VirTis Freeze Dryer (Model FFD-42, VirTis Inc., New York,

USA) and allowed to freeze dry for 72 h. The gum was then blended into powder form

using an Oster® 10 speed blender (Model 6640-33, Jarden Corporation, New York,

USA).

3.2.3. Food matrix preparation

Hydrocolloids were dissolved with the addition of sodium caseinate and waxy

corn starch. The final composition of the non-dietary fibre components was 4% w/w

starch and 6% w/w sodium caseinate. Concentrations of hydrocolloids were varied in

order to normalize the apparent viscosities from 50 to 100 s-1

, as this is the shear rate

range that exists during digestion (Borwankar, 1992). All solutions were prepared

separately by dissolving weighed quantities into distilled water at room temperature.

The solutions were left to stir until fully dissolved and then placed inside a water bath

62

at 70oC for 10 min, allowing for maximum hydration and thickening. All solutions

were left to cool at 4oC for 24 h prior to experimentation. The final concentrations of

the solutions are listed in Table 3-1.


gum (XG)-, guar gum (GG)-, soluble flaxseed gum (SFG)-, and soluble soy

polysaccharide (SSPS)-fortified solutions.

CS

XG

GG

SFG SSPS

Fibre 0 4 2 6.5 25

Sodium caseinate 6 6 6 6 6

Starch 4 4 4 4 4

Water 90 86 88 83.5 65

Total 100 100 100 100 100

3.2.4. Viscosity measurements

Solution viscosities were measured at 37±0.1 oC over a shear rate range of

30—200 s-1

, using a controlled-stress rheometer (AR 2000, TA Instruments, New

Castle, DE, USA) equipped with a 60 cm, 4o cone-plate geometry and using a

truncation gap of 50 µm. The Power Law Model (Eq. (1)) was used to analyze the

flow curves and estimate the K and n values

𝜎 = 𝐾 × �̇�𝑛 (1)

where �̇� is shear rate, n is flow behaviour index, and K is consistency index.

In the equation, shear stress (σ) is a function of the shear rate, the consistency index,

and a dimensionless exponent (n) that indicates closeness to Newtonian behaviour.

The exponent will be less than 1 for non-Newtonian shear-thinning fluids and will

equal 1 for Newtonian fluids. The consistency index is proportional to viscosity.

63

3.2.5. In vitro digestion

A two-step in vitro digestion procedure, mimicking gastric and small intestinal

digestion, was adapted from a study that developed a model for the determination of

bioavailability of soil contaminants (Oomen, Rompelberg, Bruil, Dobbe, Pereboom &

Sips, 2003). 7.0 mL of SGF [0.2%NaCl (w/w) in 0.7% HCl (w/v)] was added to an

Erlenmeyer flask containing 15 g of sample and 4 (1 cm2, 2.1 g) glass balls that were

used to induce churning and agitation. A final concentration of 3.2 mg mL-1

of pepsin

was dissolved in SGF, as described by Malaki Nik, Wright & Corredig (2010), and a

final pH of 1.8 ±0.1 was obtained. The mixture was incubated in a shaking water bath

(Thermo Scientific, Marietta, OH, USA) at 37oC, at a speed of 175 rpm, for 1 h.

Following gastric digestion, 4.6 mL simulated bile fluid (SBF), containing 8 mg mL-1

bile salts, 14 mL simulated intestinal fluid (SIF), pH 7.8 ±0.1, containing 5 mg mL-1

pancreatin dissolved in 0.5M sodium phosphate buffer, and 2.9 mL of

amyloglucosidase (112 U/mL) were added to attain a final physiological ratio of

1:2:0.5 for gastric juices, small intestinal juices, and bile, respectively (Guyton & Hall

2010). The mixture was placed back in the shaking water bath and simulated

intestinal digestion proceeded at 37oC for 3 – 4 h.

3.2.6. Glucose release measurements

Following gastric digestion, the digesta was transferred to 23 cm dialysis tubes

(Sigma-Aldrich, St. Louis, MO, USA), with a molecular weight cutoff of 2000 Da and

a width of 32 mm. The tubes were then filled with SIF, SBF, and amyloglucosidase to

mimic the small intestinal step. Two glass balls were added and the tubes were

subsequently tied off and placed inside a beaker filled with 450 mL sodium phosphate

64

buffer. Simulated intestinal digestion proceeded for 180 min with gentle agitation. The

dialysis tubes were manually inverted every 10 min to allow for sufficient mixing.

Aliquots of 0.1 mL were pipetted at 0, 10, 20, 30, 40, 50, 60, 90, 120, 150, and 180

min, replacing with 0.1 mL buffer after each withdrawal. The glucose concentration in

the dialysate was measured using the Megazyme D-glucose (GOPOD Format kit)

following the manufacturer’s instructions (Megazyme International Ireland ltd.,

Wicklow, Ireland). Absorbance was measured using a spectrophotometer (Beckman

DU 7400) at 510 nm.

3.2.7. Statistical Analysis

All experiments were performed in triplicate. Nonlinear regression analysis

was used to formulate a mathematical model of the flow behaviour data. Analysis of

variance (ANOVA) was used to determine means and standard deviations from

replicate measurements of viscosity, viscosity following in vitro digestion, and

available glucose (GraphPad Prism Software Inc., La Jolla, CA, USA). Difference test

was evaluated by using Tukey’s Multiple Comparison testing. The significance level

was set at p<0.05.

3.3.Results and discussion

3.3.1. Effect of in vitro digestion on solution viscosity

Flow curves of the four solutions before simulated digestion, after simulated

gastric digestion and following simulated intestinal digestion are shown in Fig. 3-1, 3-

2, and 3-3, respectively. All solutions behaved as pseudoplastic non-Newtonian fluids,

with the exception of SSPS after in vitro digestion, which behaved more like a

65

Newtonian fluid. The addition of the four soluble fibres to sodium caseinate – starch

suspensions led to substantial increases in solution viscosity, with no significant

differences observed between their flow behaviours at a shear rate range of 20 – 60 s-1

.

Figure 3-1. Viscosity of guar gum-, xanthan gum-, soluble soy polysaccharide

(SSPS)-, and flaxseed gum- fortified solutions, as a function of shear rate (s-1

).

After adding gastric secretions, the viscosity of all solutions was reduced, however the

degree of reduction varied with xanthan gum demonstrating the greatest resistance to

losses in viscosity and SSPS the least, as shown in Fig 3-2.

66



)

following 1h simulated gastric digestion.

In vitro digestion influences flow behaviour by depressing the flow curve, suggesting

that SSPS-containing solutions may be susceptible to the presence of hydrolytic

enzymes and/or bile. SSPS contains 66% dietary fibre, 9.2% protein, and 8.6% ash

(Maeda, 2000). The protein fraction, which in our present study was measured by the

dumas method to be 8.9%, may be susceptible to hydrolysis by human digestive

enzymes, such as pepsin and trypsin (Nakamura, Hirokazu & Corredig, 2006;

Fafaungwithayakul, Hongsprabhas & Hongsprabhas, 2011), which may have

accounted for this viscosity reduction. Fig 3-3 illustrates that upon completion of the

simulated small intestinal phase, the trend was again observed, where xanthan gum

was able to retain a substantial amount of its initial viscosity in comparison to the

other 3 gums.

67



),

following a two-stage in vitro digestion.

Xanthan gum consists of a linear 1, 4 – linked β-D-glucose backbone, with a

trisaccharide side chain at alternating glucose units, as well as a 1, 4 – linked

glucuronic acid at the terminal mannose. The side chains wrap around the cellulose-

like backbone, thereby inducing a helical-like structure that inadvertently minimizes

the likelihood of depolymerisation (Sworn & Monsanto, 2000). Evidently, the primary

structure of xanthan gum plays a protective role against hydrolysis resulting from the

various digestive processes. The resilient nature of xanthan gum has been previously

reported on, where it was shown that it maintained measurably higher apparent

viscosities at 50 and 100 s-1

in comparison to other polysaccharides that were

subjected to the same conditions (Mazza & Biliaderis, 1989), and higher nonlinear

regression parameters than both soluble and insoluble fibres (Dikeman et al., 2006).

Proteolytic and amylolytic hydrolysis, following in vitro digestion, of the protein and

68

starch components inside the solutions reduced their subsequent viscosities resulting

in flow behaviour nearly identical to when the gums were prepared in the absence of

the non-fibre components, as shown in Fig. 3-4, 3-5, 3-6, and 3-7.

Figure 3-4. Viscosity of guar gum and guar gum-fortified solution containing protein

and starch, following simulated in vitro digestion (curves are overlapping).

Figure 3-5. Viscosity of xanthan gum and xanthan gum-fortified solution containing

protein and starch, following simulated in vitro digestion.

69

Figure 3-6. Viscosity of flaxseed gum and flaxseed gum-fortified solution containing

protein and starch, following simulated in vitro digestion.

Figure 3-7. Viscosity of soy soluble polysaccharide (SSPS) and SSPS-fortified

solution containing protein and starch, following simulated in vitro digestion.

Table 3-1 shows Power Law equation parameters for the four starch- and

caseinate-containing suspensions. The flow behaviour indices demonstrate that all

solutions acted as pseudoplastic non-Newtonian fluids, both before and after in vitro

digestion, however an increase in dependence on shear rate was observed after the 4 h

70

simulated digestion. The decrease in K values indicates that after exposing all

solutions to simulated digestion, a drop in viscosity ensues.

Table 3-2. Consistency index (K) and flow behaviour index (n) from the Power Law

model of guar gum-, xanthan gum-, flaxseed gum-, and soluble soy polysaccharide

(SSPS)-fortified solutions before and after in vitro digestion.

Solution Before digestion After digestion

K (Pa.s) n K (Pa.s) n

Guar gum 178b 0.090 4.44

b 0.330

Xanthan gum 316a 0.112 27.9

a 0.160

Flaxseed gum 182b 0.240 0.745

c 0.690

SSPS 56.3c 0.570 0.0498

c 0.960

Values with different letters in the same column differ significantly (p<0.05)

In conjunction with our previous findings (Fabek, 2011), xanthan gum is able to

retain more of its viscosity in comparison to the remaining fibres, across a shear rate

range of 30 to 200 s-1

(p<0.05), as indicated by a higher consistency index shown in

Table 3-1. The guar gum-containing suspension was also able to retain some

measurable viscosity following in vitro digestion, indicated by the second highest K

value. Both flaxseed gum, and SSPS were once again on the lower end of viscosity,

with SSPS showing the lowest resilience to in vitro digestion (p<0.05). The

subsequent variability in viscosity retention between the fibre types may reflect their

capacity to hinder glucose diffusion.

71

3.3.2. Glucose diffusion during in vitro digestion

The amount of glucose liberated throughout 3 h simulated small intestinal

digestion, as a function of fibre-type, is shown in Fig. 3-8.


intestinal digestion of control (no fibre), soy soluble polysaccharide (SSPS)-, flaxseed

gum-, guar gum-, and xanthan gum-fortified solutions.

As time proceeded beyond the first 60 min of digestion the impact that the various

gums had on attenuating the diffusion of glucose into the dialysate became more

pronounced. All gums demonstrated an ability to depress the glucose response curve

in comparison to the control. However, the solution prepared with xanthan gum

showed the highest attenuation in glucose mobility, in comparison to other gums. This

can be seen by comparing the data shown in Fig. 3-8 and Table 3-2; the latter displays

the apparent viscosity of the digesta of the various solutions at a shear rate of 50 s-1

. It

has been previously reported that shear rates of 10 – 100 s-1

are found during the

various processes of digestion (Steffe, 1996) and the shear rate along the brush border

is on the lower end of that range, which is why 50 s-1

was chosen for comparison here.

72

Table 3-3. Apparent viscosity (Pa.s) at 50s-1

of guar gum-, xanthan gum-, flaxseed

gum-, and soluble soy polysaccharide (SSPS)-containing solutions after 4 h in vitro

digestion.

Solution Apparent viscosity (Pa.s) at 50 s-1

Guar gum 0.32±0.020

Xanthan gum 1.0±0.06

Flaxseed gum 0.22±0.060

SSPS 0.042±0.010

In addition, guar gum showed a lower ratio of diffused glucose, however, with the

exception of SSPS, this difference was not significantly different from the others

(p>0.05). Table 3-3 quantitatively demonstrates this by displaying the Normalized

Sugar Diffusion (NSD) values for each treatment, which were calculated using

equation 2.

𝑁𝑆𝐷 =𝐷𝐼𝐹𝐹 (𝑓𝑡)

𝐷𝐼𝐹𝐹(𝑐𝑡) (2)

where DIFF(ft) and DIFF(ct) are the amount of glucose diffused for the fibre and

control solutions, respectively at time (t).

73

Table 3-4. Normalized Sugar Diffusion (NSD) of solutions mixed with fibre

compared with the control

Solution 20 min 40 min 60 min 90 min 120 min 150 min 180 min

Control 1.0d

1.0d

1.0c

1.0d

1.0d

1.0d

1.0d

Guar gum 0.52a, b

0.52a, b

0.57a, b

0.54a, b

0.53a, b

063a, b

0.56a, b

Xanthan gum 0.49a

0.43a

0.46a

0.46a

0.43a

0.56a

0.44a

SSPS 0.75b, c

0.64b, c

0.74b

0.79c

0.74c

0.81c

0.79 c

Flaxseed gum 0.85c

0.79c

0.74b

0.62b, c

0.60b, c

0.71b, c

0.66b, c


Lower NSD values indicate a greater effect on suppressing glucose diffusion.

Throughout the 3 h experiment all fibres were able to attenuate the glucose response

in comparison to the control (NSD = 1). However, xanthan gum was the most

effective in comparison to the rest, whose NSD values were lower in comparison to

guar gum and significantly lower (p<0.05) than the remaining fibres and the control.

SSPS was the least effective but also the least viscous, which highlights the

relationship that fibre viscosity and glucose mobility share during digestion of starch-

containing foods.

The inclusion of a variety of viscous fibres in meals has been shown to reduce

postprandial glycaemia in both diabetic and non-diabetic patients (Chawla & Patil,

2010; Nilsson, Ostman, Preston & Bjorck, 2008; Jenkins et al., 1977; Brennan,

Derbyshire, Brennan & Tiwari, 2012). In the present study, the viscosity of all fibres

was normalized before any simulation and therefore it is their ability to preserve

viscosity that affects glucose diffusion across the membrane. Despite the variation in

viscosity for the different fibre types, all of the fibres employed in this study managed

74

to lower glucose diffusion in comparison to the control. Subsequently, there was an

inverse relationship between digesta viscosity and glucose concentration in the

dialysate, highlighted by the lower rate of glucose diffusion for xanthan gum and guar

gum. Similar to these results, researchers have demonstrated the ability of both of

these gums to slow starch digestion and lower diffused glucose in solutions containing

high amylose corn starch. However, the results did not attribute viscosity as being the

sole property responsible, as other more viscous gums in the study also allowed for

higher rates of starch digestion and glucose diffusion (Sasaki & Kohyama, 2012).

Similar to this, in the present study the ability of xanthan gum to withstand reductions

in viscosity is shown, whereby it was at least three times more viscous in comparison

to the others. However, the effect on attenuating glucose mobility was not as

pronounced, suggesting that viscosity alone may not be the sole contributing factor in

modulating postprandial glucose levels. The limitation associated with studying

viscous effects of dietary fibres, in vitro, is that it is difficult to determine the exact

cause of their suppressing effect(s). Viscosity may provide a protective barrier for the

starch granules, inhibiting enzyme hydrolysis. Increases in luminal viscosity may also

lead to reduction in diffusion of hydrolyzed glucose. Soluble fibres could also

contribute to a barrier layer at the mucosal membrane, slowing transport of glucose

across the mucosa. Evaluation of the effect of viscosity on both starch hydrolysis and

glucose release and diffusion may help discover the exact mechanism by which

viscosity hinders the postprandial glycemic response.

Fig 3-9 shows the amount of glucose permeating into the dialysate when D-

glucose was used as a replacement for starch. Fibre inclusion again attenuated glucose

75

mobility; however, the effect is not as pronounced in comparison to what was seen in

Fig 3-8, with starch acting as a substrate for amylolytic hydrolysis.


intestinal digestion of control (no fibre), soy soluble polysaccharide (SSPS)-, flaxseed

gum-, guar gum-, and xanthan gum- fortified solutions when starch was replaced with

D-glucose.

Therefore, although we have demonstrated that fibre-induced viscosity has a positive

impact on lowering glucose diffusivity, in vitro, there may also be an effect on the

hydrolysis of starch, indicating a potential multi-faceted manner by which soluble

fibres may lower postprandial glycemia.

3.4.Conclusion

Hydrocolloids (xanthan gum, guar gum, flaxseed gum, and SSPS) were prepared

at defined concentrations to create equi-viscous solutions. Their flow behaviour was

analyzed both before and after two-stage in vitro digestion. Xanthan gum showed

76

significantly higher resilience to the digestive processes, followed by guar gum. The

remaining gums demonstrated nearly a complete loss in viscosity by dilution from

gastric and small intestinal secretions, simulating physiological conditions. All gums

were also mixed with starch and sodium caseinate, allowing for a food-like matrix to

form. Following in vitro digestion, starch and protein were hydrolyzed and the

ensuing viscosity was attributed only to the individual gums. Xanthan gum again

showed significantly higher apparent viscosities across a range of shear rates,

followed by guar gum. Subsequently, the amount of glucose that was available

following in vitro digestion in dialysis tubes was measured and compared to the

control. All gums lowered glucose concentration, with xanthan gum being the most

effective. These results suggest that digesta viscosity of soluble fibres depends on

their ability to resist changes during digestion in addition to their initial viscosity or

concentration in solution or in a food matrix. Subsequently, this affects the capacity

by which they affect glycemic levels following in vitro digestion. These findings

suggest that the inclusion of more resilient gums, such as xanthan gum, might be an

effective strategy in lowering postprandial glycemia in humans.

77

4. Effect of soluble fibre inclusion on starch hydrolysis and

glucose mobility during simulated small intestinal

digestion

Hrvoje Fabeka and H. Douglas Goff

a,


Canada

Abstract

Type II diabetes (T2D) is a disease that afflicts millions of people annually.

Subsequently, researchers are exploring potential dietary strategies that can be used to

help improve the lifestyle of individuals diagnosed with T2D. Consumption of soluble

fibre is one such strategy, which has been studied extensively for its ability to reduce

postprandial glycemia, both in human and animal studies. The exact mechanism by

which it is able to lower plasma glucose levels remains unclear; however, researchers

speculate it is due to the increase in digesta viscosity that arises from the consumption

of fibre-rich foods. Recently, we have reported on the ability of different soluble

fibres to retain solution viscosity after passing through a 2-stage in vitro digestion

model (Fabek, Messerschmidt, Brulport, and Goff, 2014). The aim of the present

study is to follow up on our earlier work and to analyse the rheological properties of

guar gum-, xanthan gum-, soluble flaxseed gum-, and soy soluble polysaccharide

(SSPS)-starch fortified solutions during 3-stage in vitro digestion (mimicking the

salivary, gastric and small intestinal steps). Moreover, biomarkers of starch hydrolysis

will be examined by looking at the effect(s) that fibre inclusion may have on reducing

sugar and glucose concentrations inside the simulated digesta. Lastly, a dialysis

78

system will be incorporated into the study to analyse for effects of fibre-induced

viscosity on glucose diffusion.

Rheological investigations revealed that all solutions exhibited viscoelastic behaviour.

Shear-thinning behaviour was evident throughout all 3 stages of in vitro digestion,

with the exception of SSPS and flax solutions, which showed Newtonian-like

behaviour at the end of the simulated intestinal step (n value from the Power Law

model was greater than 0.9). Analysis of the small intestinal digesta demonstrated an

inverse relationship between the viscosity of the solution and the amount of reducing

sugars released, where the most viscous formulation (xanthan gum) led to the greatest

suppression of starch hydrolysis. This was also evident when measuring glucose

diffusion, where the glucose concentrations in the dialysate were significantly lower

for solutions containing a viscous polysaccharide. The findings of our current work

suggest a multi-faceted effect of soluble fibres in retarding glucose levels, by affecting

both the hydrolysis of starch (luminal digestion) as well as the diffusion of hydrolysis

products.

4.1.Introduction

The evolution of the human diet remains a contentious topic, yet it is widely

accepted that our ancestors mainly consumed plant-based foods rich in dietary fibre

(Teaford and Ungar, 2000). Unrefined grains, raw fruits and vegetables, which are all

sources of fibre, were commonly consumed and when combined with high physical

activity this provides a recipe for healthy living. However, technological advances led

to such things as processed foods, which ultimately reduced the amount of fibre that

would normally be present in a particular foodstuff. Subsequently, the rates of chronic

79

disease, such as type 2 diabetes, have been on the rise for decades (Adeghate,

Schattner, and Dunn, 2006) and it is no coincidence that this trend is paralleled with

reductions in consumer intake of dietary fibre (Kendall, Esfahani, and Jenkins, 2010).

As a result, the task of improving the well-being of humans today has extended

beyond health practitioners and is the responsibility of researchers to explore the most

appropriate ingredients for food formulation, food manufacturers to formulate

functional foods, and consumers to be informed about their food choices. In the early

1970’s, the dietary fibre hypothesis brought much of the awareness back to dietary

fibre research through the purported effects it has on lowering the risk of the

metabolic syndrome, cardiovascular disease, and diabetes (Burkitt and Trowell, 1977).

Many of the health benefits surrounding dietary fibre consumption are linked to

the ability of soluble fibres to thicken with digestive fluids (Eastwood and Morris,

1992) thereby promoting health benefits inside the small intestine through a

thickening effect. One of these supposed benefits includes reductions in postprandial

plasma glucose levels, which have been clinically studied for decades. In 1978,

researchers observed that viscous dietary fibres are responsible for being

therapeutically beneficial in moderating postprandial hyperglycemia (Jenkins,

Wolever, Leeds, Gassull, Haisman, Dilawari, Goff, Metz, and Alberti, 1978). This

relationship between increased luminal viscosity, which arises through soluble fibre

consumption, and the ensuing reductions in plasma glucose levels, remains to be a

fundamental explanation for decade’s worth of clinical studies focused on studying

soluble fibre functionality. Dikeman et al. (2006) indicated that foods containing

viscous fibres, such as oat bran and guar gum, may be beneficial for plasma glucose

80

and lipid profile attenuations, whereas non-viscous insoluble fibres may be better

suited to serve a laxative effect. However, it may be difficult to predict physiological

functionality by looking at solution viscosity alone, as digesta viscosity is sensitive to

changes introduced by the secretions of the GIT (Eastwood and Morris, 1992).

Moreover, other factors may be involved especially when fibres are components of a

food-matrix (Brennan, Blake, Ellis, and Schofield, 1996). The extent of starch

hydrolysis and the rate at which hydrolysis products are released may be major

determinants of the rates of increase of postprandial plasma glucose levels (Regand,

Chowdhury, Tosh, Wolever, and Wood, 2011).

In a food matrix, food structure and viscosity may influence the rate and extent

of starch hydrolysis. By increasing the viscosity of the surrounding medium, soluble

fibres may alter the accessibility of enzymes to starch granules (Brennan, 2005).

Subsequently, the release of intermediate starch hydrolysis products, such as α-limit

dextrins and other reducing sugars inside the lumen, may be attenuated. Reducing the

degradation and amylolysis of starch may provide an additional benefit and provide a

multi-faceted mechanism by which fibres are able to attenuate the glycemic response

(Regand et al., 2011). It is difficult to ascertain whether the action(s) of fibres along

the GIT are related to suppressing the hydrolysis of starch, retarding the diffusion of

glucose, or by providing other plausible effects (Sasaki and Kohyama, 2012).

Therefore, studying the activity of soluble fibres on starch hydrolysis is a necessary

step in understanding their functional properties as they may be related to reductions

in postprandial glycemia.

81

The objectives of this study were to fortify skim milk powder-starch mixtures

with soluble fibres and to investigate their rheological behaviour in solution. A second

objective was to develop a three-stage in vitro digestion model, mimicking the

salivary, gastric, and small intestinal phases of digestion, to which all solutions were

subsequently exposed. Afterwards, their viscosity was measured during each step.

Additionally, the storage and loss moduli of the fibre-fortified solutions were

examined, which describe the liquid-like and solid-like properties of a solution. A

third objective was to observe the effect that fibre fortification has on starch

hydrolysis by measuring the concentration of both reducing sugars and glucose inside

the digesta during simulated small intestinal digestion . Lastly, 5 h in vitro glucose

diffusion was measured using a dialysis system, to mimic the final stage of

carbohydrate metabolism.

4.2.Materials and methods

4.2.1. Materials

Four hydrocolloids were employed in this study, which included guar gum,

(Danisco Canada Inc., Toronto, ON, Canada), xanthan gum (Sigma Chemical Co., St.

Louis, MO, USA), soluble flaxseed gum extracted from flaxseed hulls (Fabek et al.,

2014), and DA-100 variety soy soluble polysaccharides (Fuji Oil Co. Ltd., Osaka,

Japan). Waxy corn starch, ULTRA-SPERSE®

A, was purchased from National Starch

(Bridgewater, NJ, USA). Skim milk powder was purchased from Gay Lea Food Corp.

(Mississauga, ON, Canada). Study-specific enzymes and chemicals were purchased

from the following distributors: simulated gastric fluid (SGF) from Ricca Chemical


82

monohydrate, sodium phosphate dibasic anhydrous, potassium chloride, potassium

citrate, potassium phosphate, and sodium chloride from Fisher Scientific (Fair Lawn,

NJ, USA), pancreatin from MP Biomedicals (Solon, OH, USA), α-amylase,

amyloglucosidase (Aspergillius niger), mucin from porcine stomach, sodium L-

lactate, ammonium nitrate, urea, and uric acid sodium salt from Sigma-Aldrich (St.

Louis, MO, USA), hydrochloric acid 2N solution from Fisher Scientific (Nepean, ON,

Canada), bile salts from Fisher Science Education (Hanover Park, IL, USA).

4.2.2. Food matrix preparation

Hydrocolloids were dissolved with the addition of skim milk powder (SMP)

and waxy corn starch. The final composition of the non-dietary fibre components were

4% w/w starch and 8.65% w/w SMP. Concentrations were varied in order to

normalize the apparent viscosities from 20 to 60 s-1

, as this is the shear rate range that

is reported during the process of digestion (Steffe, 1996). All solutions were prepared

separately by mixing the powders together and then dissolving weighed quantities into

distilled water at room temperature. The solutions were left to stir until fully dissolved

(to avoid the formation of lumps), covered and then placed inside a water bath at 70oC

for 10 min, allowing for maximum hydration and thickening. All solutions were left to

cool at 4oC for 24 h prior to experimentation. The final concentrations of the solutions

are listed in Table 4-1.

83


gum (XG)-, guar gum (GG)-, flaxseed gum (FG)-, and soluble soy polysaccharide

(SSPS)-fortified solutions

CS

XG

GG

FG SSPS

Fibre 0 4 3 7 20

Skim milk powder 8.65 8.65 8.65 8.65 8.65

Starch 4 4 4 4 4

Water 87.35 83.35 84.35 80.35 67.35

Total 100 100 100 100 100

4.2.3. Dynamic viscoelasticity and viscosity measurements

Study treatment rheology was measured using a cone and plate configuration

(cone radius of 60 mm, a truncation gap of 50.8 μm, and an angle of 2°) on a

controlled-stress rheometer (AR 2000; TA Instruments, New Castle, DE, USA).

Dynamic viscoelasticity (G’ and G”) was measured in a frequency range of 0.01 – 10

Hz at 37oC and constant stress (0.5 Pa). At this stress level, all solutions displayed

linear viscoelastic behaviour in the preliminary stress sweep test. Flow behaviour was

also measured by applying a continuous shear rate sweep from 10 to 200 s−1

with 30

sample points. The comparative shear rate range of 20 – 60 s−1

was selected based on

reports that this is the approximate shear rate during digestion (Steffe, 1996). All

measurements were performed in triplicate.


A three-step in vitro digestion procedure, mimicking salivary, gastric, and

small intestinal digestion, was adapted from our earlier study (Fabek et al., 2014) with

the introduction of a brief salivary stage to initiate amylolysis of starch. 5.0 mL of

84

simulated salivary fluid (see Table 4-2), containing alpha amylase (75 U mL-1

), was

added to Erlenmeyer flasks containing 15 g samples and 4 glass balls in each, which

were used to induce churning and agitation. Salivary digestion proceeded for 5 min

inside a shaking water bath (Thermo Scientific, Marietta, OH, USA) at 37oC, at a

speed of 60 rpm, mimicking the agitation speed during swallowing (Borwankar,

1992). Afterwards, 7.0 mL of SGF [0.2%NaCl (w/w) in 0.7% HCl (w/v)] was added,

containing 3.2 mg mL-1

of pepsin (pH = 1.8 ±0.1). The mixture was incubated at

37oC, at a speed of 175 rpm, for 1 h. Following gastric digestion, 4.6 mL simulated

bile fluid (SBF), containing 8 mg mL-1

bile salts, 14 mL simulated intestinal fluid

(SIF), pH 7.6 ±0.1, containing 5 mg mL-1

pancreatin dissolved in 0.5M sodium

phosphate buffer and calcium chloride solution, and 2.9 mL of amyloglucosidase (112

U/mL) were added to each solution. The mixture was placed back in the shaking water

bath and simulated intestinal digestion proceeded at 37oC for 5 h. All digestions were

performed in triplicate.

Table 4-2. Chemical composition of simulated salivary fluid1

Chemical Amount (g L-1

)

Sodium chloride (NaCl) 1.594

Ammonium nitrate (NH4NO3) 0.328

Potassium phosphate (KH2PO4) 0.636

Potassium chloride (KCl) 0.202

Potassium citrate (K3C6H5O7.H2O) 0.308

Uric acid sodium salt (C5H3N4O3.Na) 0.021

Urea (H2NCONH2) 0.198

Porcine gastric mucin 30

Lactic acid sodium salt (C3H5O3Na) 0.146

1Adapted from Mao and McClements (2012)

85

4.2.5. Reducing sugar determination

Reducing sugars were measured according to Woolnough, Monro, Brennan,

and Bird (2010). 1.0 mL aliquots of simulated small intestinal digesta were withdrawn

at 0 (before addition of pancreatin), 10, 20, 40, 60, 120, 240 and 300 min and added to

test tubes containing 4 mL absolute ethanol. The test tubes were centrifuged at 2500

rpm using an IEC Centra CL2 centrifuge (Sigma Aldrich, USA) for 4 min and aliquots

of 0.05 mL were withdrawn from each tube and then transferred into clean tubes.

Following this, 3.0 mL of dinitrosalicyclic acid (DNS), 0.6 mL 4M NaOH, and 0.6

mL DNS containing 0.5 mg/mL glucose were added to each of the tubes, which were

then covered and boiled in a water bath for 15 min (Miller, 1959). Subsequently, the

tubes were immersed in an ice bath for 10 min before transferring the mixture into

cuvettes. Absorbance was measured using a spectrophotometer (Beckman DU 7400)

at 540 nm. The amount of reducing sugars present was expressed in mg of reducing

sugar per g of sample, using the following formula: [adapted from Woolnough et al.

(2010)]

𝑠𝑎𝑚𝑝𝑙𝑒 𝑎𝑏𝑠

10𝑚𝑔𝑚𝐿 𝑔𝑙𝑢𝑐𝑜𝑠𝑒 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑎𝑏𝑠

𝑥 33.5 𝑥 10 𝑥 5/𝑜𝑟𝑖𝑔𝑖𝑛𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒 (𝑔)

The multiplication by 33.5 reflects the volume of digestive juices, 10 to reference the

abs reading of the glucose standard and 5 to account for the initial dilution step where

4 mL of absolute ethanol was added to 1 mL of digesta.

4.2.6. Glucose release determination inside simulated digesta

3.0 mL aliquots of simulated intestinal digesta were withdrawn at 0 (before

addition of amyloglucosidase), 30, 60, 120, 180, 240, and 300 min, and placed in 50

86

mL centrifuge tubes containing 30 mL absolute ethanol. The tubes were centrifuged at

4500 rpm for 15 min. The glucose concentration inside the supernatant was

determined using the GOPOD enzyme assay kit, according to the manufacturer’s

instructions (see Chapter 3). The degree of starch hydrolysis, as measured by the

amount of glucose released, was calculated using equation 4-1 (Hasjim, Lavau,

Gidley, and Gilbert, 2010):

% 𝑠𝑡𝑎𝑟𝑐ℎ ℎ𝑦𝑑𝑟𝑜𝑙𝑦𝑠𝑒𝑑 =(𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑔𝑙𝑢𝑐𝑜𝑠𝑒 𝑥 0.9)

𝑑𝑟𝑦 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑠𝑡𝑎𝑟𝑐ℎ 𝑖𝑛 𝑠𝑜𝑙𝑢𝑡𝑖𝑜𝑛 𝑥 100% (1)

4.2.7. Glucose diffusion inside a dialysis system

Glucose measurements were done according to the protocol described earlier

(Fabek et al., 2014). Briefly, dialysis tubes were cut to 9 cm in length and were

hydrated in deionized water for 30 min and subsequently washed off to remove any

glycerol that may have adhered to the tubes. Simulated intestinal digestion was carried

out inside the dialysis system to which two glass balls were added and the tubes were

subsequently tied off and placed inside a beaker filled with 450 mL sodium phosphate

buffer. The dialysis tubes were manually inverted every 10 min to allow for sufficient

mixing. Aliquots of 0.1 mL were pipetted at 0, 10, 20, 30, 40, 50, 60, 90, 120, 150,

180, 210, 240, 270, and 300 min, replacing with 0.1 mL buffer after each withdrawal.

The glucose concentration in the dialysate was measured using the GOPOD enzyme

assay kit and the absorbance was measured at 510 nm.

4.2.8. Statistical Analysis

All experiments were performed in triplicate. Nonlinear regression analysis

was used to develop a working model for the flow curve data. The Power Law model

87

was used to analyze the flow curves and estimate the flow behaviour (n) and

consistency (K) indices, as described by Fabek et al., 2014. Linear regression was

used to attain slopes of the reducing sugar release and glucose diffusion lines, which

were used for comparison with apparent viscosity at 50 s-1

during simulated small

intestinal digestion. Analysis of variance (ANOVA) was done to determine means and

standard deviations from replicate measurements of all experiments using GraphPad

software (GraphPad Prism Software Inc., La Jolla, CA, USA). Difference test was

evaluated by using Tukey’s Multiple Comparison testing. The significance level was

set at p<0.05.

4.3.Results and discussion

4.3.1. Study treatment rheology during in vitro digestion

Figures 4-1, 4-2, 4-3 and 4-4 show the storage (G’) and loss moduli (G”) for

the fibre-fortified treatments in solution, following simulated gastric digestion, and

after 5 h in vitro digestion.

88

Figure 4-1. Storage (G') and loss moduli (G") of xanthan gum-fortified treatments –

that is (a) in solution, (b), following simulated gastric digestion, and (c) at the end of

3-stage in vitro digestion – as a function of frequency with an oscillating stress of 0.5

Pa.

89

Figure 4-2. Storage (G') and loss moduli (G") of guar gum-fortified treatments – that

is (a) in solution, (b), following simulated gastric digestion, and (c) at the end of 3-

stage in vitro digestion – as a function of frequency with an oscillating stress of 0.5

Pa.

90

Figure 4-3. Storage (G') and loss moduli (G") of flaxseed gum-fortified treatments –

that is (a) in solution, (b), following simulated gastric digestion, and (c) at the end of

3-stage in vitro digestion – as a function of frequency with an oscillating stress of 0.5

Pa.

91

Figure 4-4. Storage (G') and loss moduli (G") of soluble soy polysaccharide-fortified

treatments – that is (a) in solution, (b), following simulated gastric digestion, and (c)

at the end of 3-stage in vitro digestion – as a function of frequency with an oscillating

stress of 0.5 Pa.

92

All solutions demonstrated viscoelastic behaviour with a measurable G’ throughout

the frequency sweep (Figures 4-1a, 4-2a, 4-3a and 4-4 a). The addition of the different

hydrocolloids to the solution resulted in the formation of a gel network, which was

observed in a previous study where researchers formulated puddings with both SSPS

and flaxseed gum (Au et al., 2013). Moreover, it has been shown that by adding

xanthan gum at concentrations above 0.5% to starch solutions there is a measurable

increase in the G’ thereby demonstrating a dependence of solution rheology on the

concentration of xanthan gum that is added (Sasaki and Kohyama, 2011). Both

xanthan gum and guar gum have been shown to increase the G’ values of tapioca

starch solutions (Chaisawang and Suphantharika, 2005). Interestingly, in our study G’

values remained higher than G” following simulated gastric digestion, with the

exception of flax-fortified solutions, which nonetheless still had a measurable elastic

component, as illustrated in Fig 4-3 b. This illustrates that all four fibre-fortified

solutions were able to maintain solid-like behaviour despite the simulated secretions

of both the salivary and gastric phases of the present in vitro model. However, both G’

and G” were measurably higher for xanthan- and guar-fortified solutions in

comparison to SSPS and flax, which reflects a greater extent of polymer

entanglements and a higher viscosity for those 2 solutions. Following simulated small

intestinal digestion, both flax- and SSPS-solutions showed complete loss of solid-like

behaviour made evident by the substantial reductions of the elastic modulus.

Moreover, the solutions exhibited substantially lower viscous components in

comparison to both guar gum and xanthan gum. Evidently, the entanglements or

interactions responsible for promoting structure in the flax- and SSPS- solutions were

93

lost following the prescribed in vitro digestion protocol. On the other hand, guar gum

and xanthan gum were able to retain their viscoelastic behaviour as both G’ and G”

were measurably higher in both solutions. Simulated digestion had a seemingly lesser

effect on the viscoelastic properties of xanthan gum-fortified solutions, with larger

values of G’ and G” in comparison to guar gum. This is in conjunction with an earlier

study where Chaisawang and Suphntharika (2005) found that tapioca starch solutions

fortified with xanthan gum had higher values for both G’ and G” in comparison to

guar gum. They attributed their findings to the lower molecular weight possessed by

guar gum. In order to make quantitative comparisons, G’, G”, and tan δ (relating the

energy lost to the energy stored, essentially the ratio of G”/G’) taken at 1 Hz are

shown in Table 4-3.

Table 4-3. Mean values for viscoelastic measurements storage (G') and loss (G")

moduli and tan δ measured at 1 Hz of xanthan gum-, guar gum-, flaxseed gum, and

soluble soy polysaccharide-fortified solutions

Solution G’ (Pa) G” (Pa) Tan δ

Xanthan gum 26.3a

5.55a

0.213a

Guar gum 6.91b

7.99a

1.15b

Flaxseed gum - 0.231b

18.9c

Soluble soy polysaccharide - 0.0993b

153d


Among the different solutions, G’ decreased in the order of xanthan gum > guar gum

> flaxseed gum > SSPS, which suggests that the extent of structure collapse due to

simulated digestion proceeds in the inverse order. The tan δ value is a measure of a

solutions physical behaviour at a given frequency, where lower tan δ values (<1)

indicate solid-like behaviour and higher values (>1) are seen in concentrated

94

solutions, and much higher values are observed in dilute solutions. Xanthan gum had a

significantly lower value for tan δ in comparison to the other 3 solutions,

demonstrating its ability to retain solid-like behaviour despite the secretions of the

simulated 3-stage digestion model. Moreover, it was the only solution that possessed a

tan δ of < 1, which is characteristic of a viscoelastic gel (true gels have tan δ values <

0.1) (Kwong, Wolever, Brummer, and Tosh, 2013). This may be attributed to the

viscoelastic properties of xanthan gum, which when placed in solution with starch

creates a weak three-dimensional network through the association of the

aforementioned helical structure and ordered chain segments, which was also reported

in earlier studies (Kim and Yoo, 2006; Doublier and Cuvelier, 1996). The other 3

treatments all behaved as solutions (tan δ > 1), however guar gum demonstrated a

much greater resistance to change, as indicated not only by a comparably lower tan δ

value (1.15, indicating the presence of a concentrated solution) but also much higher

values for G’ and G”. These results demonstrate that although all four gums employed

in the study are able to impart solid-like structures when combined with SMP and

starch, the ensuing rheological properties of the solutions following simulated

digestion are altered and become inherently dependent upon the unique

physicochemical makeup of the individual gums. Both xanthan gum and guar gum

exhibited the greatest capacity to resist changes in structure and maintained the

highest viscous component of the four treatments, with the former maintaining solid-

like behaviour.

Figure 4-5 shows apparent viscosity as a function of shear rate (10 to 200 s-1

)

for all solutions, including the control.

95

Figure 4-5. Flow behaviour of control (no gum), xanthan gum-, guar gum-, soy

soluble polysaacharide (SSPS)-, and flaxseed gum-fortified solutions across a shear

rate range of 10 to 200 s-1

.

At 20 s-1

the control solution had an apparent viscosity of 0.218 ± 0.032 Pa.s. This was

increased to 6.570 ± 0.592 Pa.s with the addition of 4% xanthan gum, 6.231 ± 0.192

Pa.s with the addition of 3% guar gum, 5.305 ± 0.317 Pa.s with the addition of 7%

flaxseed gum, and 6.239 ± 0.804 Pa.s with the addition of 20% SSPS. There was no

difference between the apparent viscosities of the fibre-fortified solutions at 20 s-1

(p>0.05). All solutions, with the exception of the control, behaved as pseudoplastic

non-Newtonian fluids, demonstrating a strong dependence on shear rate as shown by

the behaviour index values listed in Table 4-5. The viscosity of the xanthan gum

fortified solution demonstrated the greatest shear-thinning behaviour at the prescribed

shear rate range (lowest n value). This was observed in earlier studies working with

wheat starch-xanthan gum (Sajjan and Rao, 1987), corn starch-xanthan gum

(Sudhakar, Singhal, and Kulkarni, 1996) and rice starch-xanthan gum mixtures (Kim

and Yoo, 2006) and is likely due to the rigid, helical structure and high molecular

96

weight of xanthan gum thereby making it more responsive to a shear field in

comparison to polymers with random-coil conformations (Urlacher and Noble, 1997).

This unique shear-thinning behaviour makes xanthan gum a popular ingredient in

many food products. Figure 4-6 shows the effect that simulated salivary, gastric, and

small intestinal digestion has on the flow behaviour of all solutions.

Figure 4-6. Flow behaviour of control (no gum), xanthan gum-, guar gum-, soy

soluble polysaacharide (SSPS)-, and flaxseed gum-fortified solutions - that follow

simulated, (a) salivary, (b) gastric, and (c) small intestinal digestion, across a shear

rate range of 10 to 200 s-1

.

97

As expected, the viscosities decrease as the solutions pass through the 3-stage in vitro

digestion. Upon completion of the simulated small intestinal step, the flow behaviour

of both flaxseed gum- and SSPS-fortified solutions showed substantial reductions,

indicated by their respective curves in Figure 4-6 (c) and consistency index values

listed in Table 4-4.

Table 4-4. Consistency index (K, Pa.s) values of the Power Law model for control,

xanthan gum-, guar gum-, soy soluble polysaccharide (SSPS)-, and flaxseed gum-

fortified treatments in solution, following simulated salivary (SP), gastric (GP) and

small intestinal (SIP) phases of digestion.

Solution

SP

GP

SIP

Control 1.04d

0.943c

0.0240c

0.00100b

Xanthan gum 101a

80.1a

91.4a

50.1a

Guar gum 26.6b

26.8b

12.2b

3.77b

SSPS 13.8c

3.59c

0.406c 0.0400

b

Flaxseed gum 17.5b, c

5.39c

0.978c

0.193b


Pseudoplastic fluids can be adequately described using the aforementioned Power

Law equation (Steffe, 1996), which is used to calculate a consistency index or a

constant proportional to viscosity, as described in Chapter 3. Guar gum retained a

greater sum of its initial viscosity following in vitro digestion in comparison to both

flaxseed gum and SSPS; however no significant difference was observed in the

viscosity constant shown in Table 4-4 or in the flow behaviour (Figure 4-6 c) when

compared with the control, flax- or SSPS-solution at shear rates > 75 s-1

. On the other

hand, xanthan gum retained a significant amount of its viscosity throughout the

prescribed shear rate range (p<0.05), which corresponds with our earlier work in

Chapter 3 where we reported on the tremendous resilience of xanthan gum towards

98

the digestive secretions mimicked in a 2-stage in vitro protocol (Fabek et al., 2014).

Overall, during simulated small intestinal digestion, xanthan gum was the only

solution whose viscosity differed from the control and the other treatments (p<0.05),

as indicated by the viscosity constant listed in Table 4-4. Moreover, although viscosity

of xanthan gum solutions remained highly dependent on shear rate, there was a

reduction in the extent of shear-thinning following simulated small intestinal

digestion, as shown by an increase in the n value in Table 4-5. As the exponent (n)

value approaches 1, solutions become less dependent on shear rate and effectively

behave more as Newtonian fluids (Dikeman et al., 2006), which was seen in the

control, flaxseed gum- and SSPS-fortified solutions following in vitro digestion.

Table 4-5. Behaviour index (n) values of the Power Law model for control, xanthan

gum-, guar gum-, soy soluble polysaccharide (SSPS)-, and flaxseed gum-fortified


intestinal phases of digestion.

Solution

SP

GP

SIP

Control 0.474d

0.295b

0.801c

0.971d

Xanthan gum 1.93x10-8 a

3.46x10-8 a

0.022a

0.033a

Guar gum 0.505b

0.418c

0.494b

0.396b

SSPS 0.736d

0.680d

0.835c

0.977d

Flaxseed gum 0.586c 0.622

d 0.731

c 0.848

c


The slight change in shear-thinning behaviour of xanthan gum as digestion time

proceeds may be attributed by the dilution effect that the simulated secretions have on

the structural conformation of xanthan in solution. Other studies have shown that

lower concentrations of xanthan gum result in higher behaviour index values (Kim

and Yoo, 2006) and therefore lower dependence on shear rate. In the present study,

99

despite comparable viscosities between all fibre-fortified solutions prior to simulated

digestion (Figure 4-5) xanthan gum was able to retain the greatest amount of its initial

viscosity in comparison to the other three gums (Figure 4-6 c), which may elicit

beneficial physiological responses, such as attenuations in glycemia, where viscosity

is believed to play an important role.

4.3.2. Reducing sugar release during simulated small intestinal digestion

Figure 4-7 illustrates the effect of soluble fibre inclusion on reducing sugar

release (RSR) from solution during 5 h simulated small intestinal digestion.

Figure 4-7. Reducing sugars released, expressed as mg/g sample of control (no fibre),

soy soluble polysaccharide (SSPS)-, flaxseed gum-, guar gum-, and xanthan gum-

fortified solutions, during 5 h simulated intestinal digestion.

The results demonstrate that reductions in RSR are achieved by incorporating

different dietary fibres into starch-containing solutions. There is an effect not only in

the amount of total reducing sugars released but also the rate of sugar release. For

example, the evolution of reducing sugars from 60 to 300 min is significantly slower

100

for the xanthan gum-fortified samples compared with the control sample (p<0.05).

Linear regression analysis was performed to quantitatively compare the effects that

the different treatments have on RSR by calculating the slope of each line, as shown

in Table 4-6.

Table 4-7. Calculated slopes of the lines of regression (SLREG) of reducing sugar

release for control, xanthan gum-, guar gum-, flaxseed gum, and soy soluble

polyssacharide (SSPS)-fortified treatments.

Treatment SLREG R2

Control 0.9075c 0.98

Xanthan gum 0.4218a

0.98

Guar gum 0.6953b

0.96

Flaxseed gum 0.7912b,c

0.98

SSPS 0.8778c

0.99


The linear regression model fit the data very well, as indicated by R2 values ranging

from 0.96 to 0.99. Evidently, there is variation in the effectiveness by which fibres are

able to attenuate RSR. Xanthan gum elicited the greatest effect on lowering RSR,

followed by guar gum, and both flaxseed gum and SSPS demonstrated having a lesser

effect. This may be due to reductions in starch hydrolysis caused by the increase in

digesta viscosity that the different fibres are able to promote. Another study has

shown the effect that β-glucan inclusion in breads has on RSR. The researchers

observed reductions in RSR by increasing the concentration of β-glucan and although

they could not fully explain their findings they believe the attenuations could result by

not only the diminished amount of starch that remains available for digestion but also

the increase in digesta viscosity that is generated through the inclusion of β-glucan

101

(Symons and Brennan, 2004), which is in conjunction with reports made in earlier

studies (Braeten, Wood, Scott, Riedel, Poste, and Collins, 1991; Wood, Beer, and

Butler, 2000) As shown in the previous section, fibres resist changes in rheological

behaviour variably as the solutions are subjected to the simulated digestion model.

Subsequently, the range in apparent viscosities could explain the results shown here.

Figure 4-8 plots regression relationships between the RSR data from Table 4-7 against

apparent viscosities at 50s-1

for each treatment.

Figure 4-8. Slopes of the lines of regression (SLERG) of reducing sugar release for

control, xanthan gum-, guar gum-, flaxseed gum, and soy soluble polyssacharide

(SSPS)-fortified treatments versus apparent viscosity (50s-1

) during simulated small

intestinal digestion.

There is a strong, negative correlation between RSR and digesta viscosity, where the

most viscous fibre of the study (xanthan gum) was also the most effective at lowering

the rate of reducing sugar release. However, it still remains unclear whether the effect

is through providing a physical barrier to amylolytic attack or if fibres somehow

function through other means inside the lumen. Nonetheless, the results of the present

102

in vitro study demonstrate a clear relationship between digesta viscosity and the extent

of RSR.

4.3.3. Effects of viscous fibres on glucose release during simulated small intestinal

digestion

The in vitro digestion behaviour of the different solutions, as measured by the

amount of glucose released inside the simulated digesta, is illustrated in Figure 4-9.

Figure 4-9. Percentage of starch hydrolysis during 5 h simulated small intestinal

digestion of control (no fibre), soy soluble polysaccharide (SSPS)-, flaxseed gum-,

guar gum-, and xanthan gum-fortified solutions (measured by the concentration of

glucose).

Nearly half of the starch was converted to glucose following the 5 h digestion step,

with the majority of hydrolysis taking place in the first 2 h. A similar trend was

observed when a starch-containing pig diet was exposed to an in vitro digestion

103

model. The researchers of that study observed less than 50% of the starch to be

hydrolysed at 5 h and reported that it took 25 h of simulated digestion to hydrolyse

nearly 98% of the starch (Hasjim et al., 2010). Moreover, studies looking at the

kinetics of starch digestion have noted that non-digested starch would be present even

after 16 h of in vitro digestion (Mahasukhonthachat, Sopade, and Gidley, 2010). A

recent study reported on the antagonistic relationship that exists between endo- and

exo-acting enzymes. The researchers indicated that in the presence of α amylase,

amyloglucosidase binds to longer starch chains (digestion products of amylolyis) with

less efficiency and subsequently allows for ineffective hydrolysis of maltose and

maltotriose (Zhang, Dhital, and Gidley, 2013), which may explain why even in the

control treatment there is undigested starch remaining after 4 h in vitro digestion.

Interestingly, in our study all fibre-fortified solutions showed substantial reductions in

the amount of glucose being released inside the simulated digesta. Table 4-8 lists the

percentage of starch that was hydrolysed in each treatment following of in vitro

digestion.

Table 4-8. Percentage of starch hydrolysed (as measured by the amount of glucose

released) after in vitro digestion of control (no gum), xanthan gum-, guar gum,

flaxseed gum, and soy soluble polysaccharide (SSPS)-fortified solutions.

Treatment % starch hydrolysed

Control 40.0a

Xanthan gum 2.80b

Guar gum 5.20b

Flaxseed gum 12.4b

SSPS 16.0b


104

The most viscous system (xanthan gum) led to the most pronounced reduction in

starch hydrolysis; however, this effect is not significant in comparison to the other

fibre treatments, contrasting what was observed in the reducing sugar analysis. This

observation may be explained by the dualistic action that fibres may exert during

starch digestion. As discussed in Section 2, in vivo glucose release occurs at the

basolateral membrane via the action of glycolytic enzymes embedded there. However,

in order for starch to be broken down into glucose, endohydrolases (α-amylase) must

first have access to the starch (luminal digestion) and the subsequent products must

then be hydrolysed by exohydrolases (amyloglucosidase) to produce glucose

(membrane-level digestion). We have observed that both xanthan gum and guar gum

are able to lower RSR thereby providing a slower release of sugars. Subsequently,

there could be an additional concurrent mechanism by which fibres are able to

attenuate the glycemic response by also affecting the hydrolysis of luminal digestion

products (reducing sugars). In the present study, even the low viscosity fibres (SSPS

and flaxseed gum) are able to reduce the amount of glucose in the digesta, which may

be due to the viscosity of the solutions or by increasing mass transfer resistance

(Singh, Dartois, and Kaur, 2010). A recent study reported that both SSPS and flaxseed

gum inclusion in dairy products resulted in blunted postprandial glucose metabolism.

In their study, the reported viscosity in solution was approximately 75 mPa.s, for both

SSPS and flaxseed gum (Au et al., 2013), whereas in our present study simulated

digesta viscosity ranged from 40 to 120 mPa.s, for SSPS and flaxseed gum,

respectively. Moreover, an earlier study reported that viscosities as low as 10 mPa.s

(in solution) are able to modulate postprandial glucose levels with reductions in peak

105

plasma glucose reaching >12% (Wood, Braaten, Scott, Riedel, Wolynetz, and Collins,

1994). It is therefore plausible that fibres, even at low viscosities, have the potential of

lowering glucose levels by means of impacting the complete breakdown of starch;

however, this requires further investigation. Nonetheless, soluble fibre inclusion in the

present study led to pronounced reductions in starch hydrolysis, with the results

demonstrating an inverse relationship between digesta viscosity and glucose release.

However, it is difficult to ascertain whether the fibres are decreasing the rate of

glucose release through the inhibition of enzymes, or whether increases in digesta

viscosity result in limiting diffusion of amylolytic products and create impairment of

mixing of digestive contents.

Additionally, the solutions were tested for the ability of soluble fibres to retain

glucose using the in vitro diffusion assay described earlier (Fabek et al., 2014). Figure

4-10 shows the amount of glucose released through the dialysis membrane during 5 h

simulated small intestinal digestion.

106

Figure 4-10. In vitro diffusion of glucose during 5 h simulated small intestinal

digestion of control (no fibre), soy soluble polysaccharide (SSPS)-, flaxseed gum-,

guar gum-, and xanthan gum-fortified solutions.

In vitro digestion of the control solution led to significantly greater release of glucose

during the first 2 h of digestion. Interestingly, even the low viscosity fibres were

observed to exert impedance on the rate of diffusion during the first 2 h. However, this

difference was not observed upon completion of the 5 h simulated digestion.

Quantitative comparisons of regression analyses are listed in Table 4-9.

107

Table 4-9. Final glucose concentrations measured inside the dialysate and calculated

slopes of the lines of regression (SLREG) of glucose release for control, xanthan gum-

, guar gum-, flaxseed gum, and soy soluble polysaccharide (SSPS)-fortified treatments

during 5 h in vitro digestion.

Treatment Final

glucose concentration (g)

SLREG R2

Control 0.515c

0.378c 0.95

Xanthan gum 0.257a

0.193a

0.99

Guar gum 0.381b

0.285b

0.96

Flaxseed gum 0.480c

0.377c

0.98

SSPS 0.482c

0.362c

0.96


Nonetheless, despite non-significant differences in the final glucose concentrations

measured in the dialysate (Table 4-9), the rate of release is more gradual for all fibre-

fortified solutions, including SSPS and flaxseed gum. However, treatments whose

viscosities were less affected by the simulated secretions allowed for significantly

greater reductions in the diffusion of glucose through the membrane, where xanthan

gum was observed to have the most pronounced effect. These findings are in

conjunction with our earlier work (Fabek et al., 2014). Increased digesta viscosity

through supplementation of soluble fibres may reduce the postprandial glucose

response by way of impeding on the diffusion of glucose (Sasaki and Kohyama,

2012). Additionally, it may be due to the ability of certain fibres to create a gel matrix

inside the GIT. As discussed previously, despite the secretions of the 3-stage in vitro

digestion model, xanthan gum maintained a solid-like network (G’>G”, tan δ<1) and

subsequently exerted the greatest reduction on glucose diffusion. Another study,

working with drinks and gels that were fortified with β-glucan, demonstrated that the

gels had a much more pronounced effect at lowering the in vitro diffusion of glucose

108

in comparison to the β-glucan-fortified drinks. However, the researchers observed the

opposite effect when measuring plasma glucose, in vivo. They explained their results

by the openness of the gel network and possible disruption of the gel matrix during

chewing, which was a condition that was absent in the in vitro portion of their study

(Kwong et al., 2013). Similarly, in our model system, in addition to promoting

viscosity it is plausible that the glucose is trapped inside the gel network, thereby

being released at a much slower rate in comparison to solutions that lack a measurable

G’, such as in the control, SSPS, and flaxseed gum treatments.

4.4.Conclusion

Starch/SMP solutions were fortified with different gums, which subsequently

increased the viscosity of the solutions and led to the formation of viscoelastic gels.

Despite similar flow behaviours in solution, after passing through the 3-stage in vitro

digestion system the rheology of the solutions changed. Both flaxseed gum and SSPS

showed complete loss of solid structure and substantial reductions in viscosity. Guar

gum exhibited slight resistance to change and maintained a measurable elastic

component and slightly higher viscosity. On the other hand, xanthan gum maintained

its solid network after a 5 h simulated digestion and had significantly higher viscosity

in comparison to the other treatments. Subsequently, an inverse relationship was

observed between viscosity and RSR, whereby xanthan gum was the most effective at

reducing starch hydrolysis. Moreover, glucose release inside the simulated digesta

was significantly reduced through fibre-fortification where even the fibres with lower

viscosity were able to affect glucose release. However, this result was less apparent

when observing the diffusion of glucose across a dialysis membrane.

109

The effects that fibre-fortified solutions have on the in vitro glycemic response

could be explained by their rheological properties and classifications inside the

simulated digesta. The presence of viscous fibres attenuates the extent by which starch

is hydrolysed, thereby limiting the release of both reducing sugars and glucose.

Moreover, having a gel network present may act to entrap glucose molecules allowing

for blunted diffusion across the membrane. Although these comparisons can be made

by means of in vitro analyses, caution must be taken when attempting to interpret

results from in vitro studies as means of explaining the physiological effects of soluble

fibres. Moreover, these effects on glycemia may not be explained by diffusion alone,

other events, such as gastric emptying, hormonal control or enzyme activity, may be

affected through fibre supplementation, which may be difficult to analyse using

current in vitro techniques. Nonetheless, as a follow up to our study further

investigations on structure-function relationships of soluble fibres will be of great

value in advancing our understanding of the impact they have on glycemic control.

Particular attention should be paid to studying the effects that fibre fortification has on

starch hydrolysis, in order to better explain the findings shown in this study.

110

5. Simulated intestinal hydrolysis of native tapioca starch:

understanding the effect of soluble fibre

Hrvoje Fabek

a and H. Douglas Goff

a,


Canada

Abstract

Starch is an energy reserve found in grains, fruits, stems, roots and tubers and is the

primary energy source in human food and animal feed. Starch structure is directly

related to digestibility, which in turn affects human health. Rapidly available starch

leads to increases in both blood glucose and insulin levels and has been studied

extensively to determine its relation to metabolic diseases, obesity and diabetes.

Soluble fibre has been shown to have a reducing effect on plasma glucose levels,

which is of particular importance due to the rapidly increasing rates in type 2 diabetes.

Researchers speculate that an increase in digesta viscosity is responsible for lowering

glucose in the blood; however, the exact mechanism remains to be elucidated. In the

present study we examined the effect that adding viscous soluble fibres would have on

starch digestibility during simulated intestinal digestion in order to better understand

the mechanism surrounding fibre functionality.

Tapioca starch, a common inclusion in many food products, was formulated with skim

milk powder to act as a control. The treatments consisted of 4 soluble fibres: xanthan

gum, guar gum, soluble flaxseed gum and soy soluble polysaccharide (SSPS) at

constant viscosity. Subsequently, all solutions were exposed to a 3-stage in vitro

111

digestion, mimicking the salivary, gastric, and small intestinal steps. During the final

stage, starch granules were isolated at varying times to observe for changes to granule

structure. Light scattering results showed that the d(4,3) values, which ranged from 16

to 20 μm, decreased as digestion proceeded. Starch degradation was also investigated

using scanning electron and light microscopy. Images from the micrographs showed

evidence of pitting and cracking along the surface of the granules extracted from the

control, flax and SSPS solutions and both endo- and exo-corrosion were observed

upon completion of simulated small intestinal digestion. The progression of these

changes was attenuated for granules that were extracted from the guar gum and

xanthan gum solutions, which we believe is linked to their rheological behaviour as

both solutions had greater viscosities inside the digesta in comparison to the other

treatments. The results of this study suggest that the glucose-lowering ability of

viscous fibres may be related to their ability to reduce the rate at which starch granules

are hydrolysed inside the lumen.

5.1. Introduction

Starch is a major component of our everyday diet as it is the main energy source

in human food. Starch is found encapsulated inside granules in a variety of plant

tissues and organs, including roots, leaves, shoots, fruits, grains, and stems (Eliasson,

2004). In grains, it is abundant in rice, maize, barley, wheat, sorghum, pulses and

tubers (Dona, Pages, Gilbert, and Kuchel, 2010). Most naturally occurring starch

granules consist of two types of polysaccharides, amylose and amylopectin. Both are

polymers consisting of α-(1-4)-D-glucose, where amylose is the smaller of the two,

with a molecular weight (MW) of approximately 106 Da, and is a primarily linear

112

polymer with few branch points along the backbone. Amylopectin on the other hand

consists of a large number of short chains bound at the reducing ends via α-1,6-

linkages, creating a much larger (MW is approximately 108 Da) branched molecule

and it makes up approximately 70% of most starch varieties (Takeda, Hizukuri, and

Juliano, 1986; Dona, Pages, Gilbert, and Kuchel, 2010). Because both polymers are α-

glucans, starch is readily digestible in humans. It is for that reason that starch

digestibility has a tremendous impact on human health. Subsequently, many

researchers have focused their studies on analysing in vitro starch digestibility in order

to better understand the link between food consumption and chronic disease.

Although starch digestion is initiated inside the mouth through the action of

ptyalin (salivary amylase), the majority of it is carried out in the small intestine where

pancreatic α-amylase hydrolyses the polymer into smaller fragments, mainly maltose,

maltotriose and α-limit dextrins that are subsequently hydrolysed to glucose through

the action of brush-border enzymes, which are exo-glucosidases that act on the non-

reducing end of the oligomers, prior to glucose being absorbed (Gray, 1992; Boron

and Boulpaep, 2009). Subsequently, starch can be classified by its in vitro digestibility

(Englyst, Kingman, and Cummings, 1992), where rapidly digestible starch (RDS) is

hydrolysed into glucose within the first 20 min, slowly digestible starch (SDS)

releases glucose between 20 and 120 min, and resistant starch (RS) escapes digestion

in the upper small intestine and is defined as the total starch minus the amount of

glucose released during 120 min of in vitro digestion (Singh, Dartois, and Kaur,

2010). Therefore, unlike RDS, which has been shown to promote metabolic

syndrome, including insulin resistance and diabetes (Byrnes, Miller, and Denyer,

113

1995), RS remains unhydrolysed as it passes through the upper GIT and consequently

has a negligible effect on the glucose response. In addition to enzymatic hydrolysis

being affected by the physicochemical properties of the starch granules themselves,

incorporation of non-starch polysaccharides, specifically soluble fibres, have been

associated with glycemic reductions.

Soluble fibres thicken when they come in contact with digestive secretions and it

is through this increase in viscosity that soluble fibres are believed to have an effect

on glycemic reductions, as shown through both in vitro and in vivo studies (Dikeman

and Fahey Jr., 2007; Brownlee, 2011). Despite accumulating research studies

indicating the ability of soluble fibres to decrease the rate of starch degradation, the

exact mechanism(s) by which they are able to do so remain unclear. Although studies

have been done that have examined how rheological properties might impact starch

hydrolysis (Sasaki and Kohyama, 2012, Fabek et al., 2014), the researchers use sugar

release as an indicator of starch digestibility.

Recently, we reported on the ability of different fibres to not only reduce the

amount of glucose that is being liberated and diffused throughout simulated digestion,

but also the impact they have on lowering reducing sugar release. The findings of our

study indicated that soluble fibre functionality, as it pertains to in vitro glycemic

reductions, may be multi-faceted through reducing both luminal digestion of starch

and the subsequent release/mobility of glucose (Chapter 4). However, the study was

conducted using quantitative analyses of starch digestion. Brennan (2005) suggested

a potential effect of soluble fibre-induced viscosity on altering starch granule

digestion. Moreover, interactions of starch with dietary fibre in pasta and cereal

114

products have been shown to reduce starch digestion thereby lowering the glycemic

response (Granfeldt and Bjorck, 1991; Brennan and Tudorica, 2008; Aravind, Sissons,

and Fellows, 2012). However, few studies have examined the rheological properties

of fibres during simulated digestion and how they relate to starch degradation and

even fewer studies examine the effects of different types of soluble fibres. Although

most human foods that contain starch are cooked (gelatinized), cooking can greatly

modify, degrade and even destroy granular structure (Dona et al., 2010).

Subsequently, most studies using fibre-enriched foods such as cereals, pastas, and

breads allow for starch gelatinization, which makes it difficult to visualize the likely

changes occurring to the microstructure of starch granules. Despite numerous research

studies looking at the effects of dietary fibres on starch digestion, it remains unclear as

to how different types of fibre are able to impact the hydrolysis of starch granules and

therefore moderate the in vitro glycemic response.

This study was undertaken as a follow up to our earlier work, which focused on

the effects of soluble fibre inclusion on reducing sugar release (RSR) and glucose

liberation. The aim of the present study is to investigate the effects that viscous fibres,

xanthan gum (XG), guar gum (GG), soluble flaxseed gum (SFG), and soy soluble

polysaccharide (SSPS) have on (uncooked) starch granules during in vitro digestion.

Granules were extracted and purified from digesta at different times of simulated

small intestinal digestion to observe for changes in particle size using light scattering.

Microscopy analyses were also performed on isolated starch granules to visualize

changes occurring in surface topography using scanning electron microscopy and light

microscopy.

115

5.2. Materials and methods

5.2.1. Materials

Four hydrocolloids were employed in this study, which included guar gum,

(Danisco Canada Inc., Toronto, ON, Canada), xanthan gum (Sigma Chemical CO., St.

Louis, MO, USA), soluble flaxseed gum extracted from flaxseed hulls (Fabek et al.,

2014), and DA-100 variety soy soluble polysaccharides (Fuji Oil Co. Ltd., Osaka,

Japan). Tapioca starch (batch number H88 80008) was purchased from Ingredion

(Bridgewater, NJ, USA). Skim milk powder was purchased from Gay Lea Food Corp.

(Mississauga, ON, Canada). Study-specific enzymes and chemicals were purchased

from the following distributors: simulated gastric fluid (SGF) from Ricca Chemical


monohydrate, sodium phosphate dibasic anhydrous, potassium chloride, potassium

citrate, potassium phosphate, and sodium chloride from Fisher Scientific (Fair Lawn,

NJ, USA), pancreatin from MP Biomedicals (Solon, OH, USA), α-amylase,

amyloglucosidase (Aspergillius niger), mucin from porcine stomach, sodium L-

lactate, ammonium nitrate, urea, and uric acid sodium salt from Sigma-Aldrich (St.

Louis, MO, USA), hydrochloric acid 2N solution and glycerol from Fisher Scientific

(Nepean, ON, Canada), bile salts from Fisher Science Education (Hanover Park, IL,

USA) anhydrous ethyl alcohol from Commercial Alcohols, the industrial and

beverage alcohol division of GreenField Ethanol Inc. (Brampton, ON, Canada).

116

5.2.2. Sample preparation

Solutions of commercial (native) tapioca starch and skim milk powder (SMP)

were prepared by dissolving weighed amounts of each in deionised water. To

investigate the effect of soluble fibre inclusion, XG, GG, SFG and SSPS were added,

according to their matched apparent viscosities at 20 s-1

(Chapter 4). The final

concentrations of XG, GG, SFG, and SSPS were 4% w/w, 3% w/w, 7% w/w and 20%

w/w, respectively, and all solutions contained the same concentration of starch (4%

w/w) and SMP (8.65% w/w). The solutions were left to stir until fully dissolved and

the control sample (no fibre) was left to stir at room temperature until experimentation

to avoid sedimentation of starch granules. All solutions were prepared at room

temperature to avoid granule swelling and amylose leaching in order to preserve the

granular structure of starch.


All treatments were digested using a three-stage in vitro digestion procedure,

mimicking salivary, gastric, and small intestinal digestion, adapted from our earlier

studies (Fabek et al., 2014; Fabek and Goff, 2015). In short, 15 g samples with 4 glass

balls (to induce churning) were preincubated with artificial salivary fluid (5.0 mL)

containing alpha amylase (75 U mL-1

) for 5 min inside a shaking water bath (Thermo

Scientific, Marietta, OH, USA) at 37oC, at a speed of 60 rpm, mimicking the agitation

speed during swallowing (Borwankar, 1992). The ensuing digest was then exposed to

7 mL simulated gastric fluid, [0.2%NaCl (w/w) in 0.7% HCl (w/v)], containing 3.2

mg mL-1

of pepsin (pH = 1.8 ±0.1). The mixture was incubated at 37oC, at a speed of

117

175 rpm, for 1 h. Following simulated gastric digestion, 4.6 mL simulated bile fluid

(SBF), containing 8 mg mL-1

bile salts, 14 mL simulated intestinal fluid (SIF), pH 7.6

±0.1, containing 5 mg mL-1

pancreatin dissolved in 0.5M sodium phosphate buffer

and calcium chloride solution, and 2.9 mL of amyloglucosidase (112 U/mL) were

added to each solution. The mixture was placed back in the shaking water bath and

simulated intestinal digestion proceeded at 37oC for 4 h. All digestions were


5.2.4. Viscosity measurements

Study treatment viscosity was measured using a cone and plate configuration

(cone radius of 60 mm, a truncation gap of 50.8 μm, and an angle of 2°) on a

controlled-stress rheometer (AR 2000; TA Instruments, New Castle, DE, USA). Flow

behaviour was measured by applying a continuous shear rate sweep from 10 to 200 s−1

with 37 sample points. The comparative shear rate range of 20 – 60 s−1

was selected

based on earlier reports that this is the approximate shear rate during digestion

(Borwankar, 1992; Steffe, 1996). Viscosity of the digesta was also measured during

simulated small intestinal digestion and represented by the consistency index (K

value) of the Power Law model (Fabek et al., 2014). All measurements were


5.2.5. Isolation and purification of starch granules for morphological study

Starch granules were isolated at varying times from the in vitro digesta using a

modified method of Hasjim, Lavau, Gidley, and Gilbert (2010). Aliquots of digesta

(30 mL) were filtered under vacuum through a screen with 500 μm openings.

Subsequently, the residue remaining on the screen was washed with sodium chloride

118

(0.1 M, 100 mL) and the filtrate was filtered again and the screen washed a second

time with sodium chloride solution. Toluene (75 mL) was added to the final filtrate

and stirred for 2 h. The mixture was then left to sit until the starch granules

sedimented and an emulsion layer of denatured protein formed on the top. The protein

and sodium chloride solution were discarded and another 75 mL toluene was added to

the starch layer and mixed for an additional 30 min. The mixture was once more left

to sit until the remaining protein was removed, which was evidenced by a clear

toluene solution following sedimentation of starch granules. The granules were

washed four times with 4 mL deionized water and 10 mL absolute ethanol. The

mixture was then transferred into centrifuge tubes and centrifuged at 4500 rpm for 10

min using a Heraeus Multifuge X1R centrifuge (ThermoFisher Scientific, MA, USA)

in order to separate the starch. The granules were collected and left to dry for 72 h at

room temperature inside a fumehood to prepare for further analysis. The protein

content of the starch was determined using the dumas method and was observed to be

0.4% ± 0.03 for all granules isolated from each of the five treatments.

5.2.6. Particle Size Analysis

Particle size analysis of isolated starch granules was carried out using a

Malvern Mastersizer Hydro 2000 SM (Malvern Instruments Ltd., Malvern WR14

1XZ, Worcestershire, UK). The granules were suspended in water and stirred at 2800

rpm. A general purpose analysis model was employed with particle refractive indices

of 1.52 and 0.001, respectively, while the refractive index of the dispersant (water)

was 1.33. Background corrections and system alignment were conducted before each

measurement. Particle size was defined as the volume weighted mean (d[4,3]).

119

5.2.7. Scanning electron microscopy (SEM)

Morphologies of digested starch granules isolated from the control (no fibre),

XG-, GG-, SFG- and SSPS-fortified solutions were analysed before and throughout

simulated small intestinal digestion. Observations were made after 0 h (before

addition of amyloglucosidase), 1 h, 2h, and 4h of amylolysis, by using a Hitachi

scanning electron microscope (S570, Nissei Sangyo Inc., Rexdale, ON, Canada) at an

accelerating potential of 20 kV. Dry starch samples were brushed onto the surface of

double-sided carbon adhesive tape mounted on an aluminum stub and subsequently

coated with 20 nm of gold in an argon atmosphere. Fifteen frames of pictures of each

sample were taken to represent the morphology of isolated starch granules.

5.2.8. Light microscopy

Starch granules isolated from in vitro digesta were suspended in 50% glycerol

(v/v) and observed under a light microscope (magnification 40x) using an Olympus

BX60 trinocular microscope (Olympus America Inc., Center Valley, PA, USA)

equipped with cellSens Entry 1.9 digital imaging software, which was used for image

capture.

5.2.9. Statistical analysis

Analytical determinations and all experiments were performed in triplicate.

Analysis of variance (ANOVA) was done to determine both means and standard

deviations and a comparison of means was performed using Tukey’s Multiple

Comparison testing at a 5% level of significance using GraphPad software (GraphPad

Prism Software Inc., La Jolla, CA, USA).

120

5.3. Results and discussion

5.3.1. Flow behaviour

The flow behaviours of the fibre-fortified treatments are shown in Figure 5-1.

The formulations were chosen to allow for equi-viscous apparent viscosities at the

chosen shear rate range of 20 – 60 s-1

, as shown in our previous work (Fabek and

Goff, 2015).

Figure 5-1. Flow behaviour of native starch/skim milk powder solutions fortified with

4% xanthan gum, 3% guar gum, 7% soluble flaxseed gum, and 20% soy soluble

polysaccharide (SSPS).

As expected, all fibre-fortified solutions behaved as non-Newtonian fluids and the

pseudoplasticity is clearly demonstrated in Table 5-1, which displays both the

consistency index (K) and flow behaviour index (n) values of the Power Law model

during simulated small intestinal digestion.

121

Table 5-1. Power Law parameters of control (no fibre) xanthan gum (XG)-, guar gum

(GG)-, soluble flaxseed gum (SFG)-, and soy soluble polysaccharide (SSPS)-fortified

treatments of the digesta isolated at the beginning (T0), after 1 h (T1), 2 h (T2), and at

the end (T3) of simulated small intestinal digestion.

Treatment Consistency index (K) (Pa.s)

IS1

T0 T1 T2 T3

Control 1.11 0.000284 0.000229

0.000119

0.000120

XG 126 77.2 53.0 35.3 20.0

GG 104 96.7 16.1 10.8 3.95

SFG 23.1 0.145 0.0829 0.0692 0.0634

SSPS 11.4 0.0444 0.0309 0.0303 0.0225

Behaviour index (n)

IS1

T0 T1 T2 T3

Control 0.476 0.818

0.855 0.969 0.980

XG 2.37x10-8

8.92x10-7

0.0515 0.0837 0.139

GG 0.0855 0.145

0.330 0.200 0.510

SFG 0.497 0.825b

0.868 0.876 0.880

SSPS 0.701 0.866

0.888 0.891 0.928

1IS=In Solution (before exposing the treatments to the in vitro digestion model)

As digestion proceeded to completion, substantial reductions in the viscosity of the

digesta of control, SFG- and SSPS-fortified treatments were observed. This is in

conjunction with our earlier work where we demonstrated a loss in viscosity due to

the hydrolysis of the starch and protein components and the dilution of the non-starch

polysaccharides (Fabek et al., 2014). Both XG and GG demonstrated the greatest

ability to resist changes in viscosity, with the former being the most resilient.

Interestingly, the reduction in digesta viscosity is progressive for both XG and GG (K

value continues to decrease from To to T4), whereas the control, SFG- and SSPS-

treatments demonstrated no significant changes in K values after 1 h of simulated

small intestinal digestion. This may be due to the ability of XG and GG to attenuate

122

starch hydrolysis. In our earlier study we observed suppressive effects of both gums

on RSR and glucose release, with XG (the most viscous fibre) eliciting the greatest

effect on the subsequent release of sugars. As a result, the progressive breakdown of

starch in the solutions could explain why there is a gradual loss in viscosity. The

ability of XG to resist rheological changes during in vitro digestion was discussed in

our earlier work (Chapter 4) and remains consistent in the present study despite the

substitution for native (non-gelatinized) starch. The behaviour index values

demonstrate that all treatments behaved as shear-thinning solutions before being

subjected to the in vitro digestion model, with a high dependence on shear rate. This

trend was less apparent as hydrolysis continued, with the control, SFG- and SSPS-

fortified solutions behaving as Newtonian fluids upon completion of simulated small

intestinal digestion (n values close to 1). Moreover, despite an increase in the n value,

the unique helical conformation of XG allowed it to remain strongly shear-thinning

throughout in vitro digestion.

5.3.2. Particle size distribution of hydrolysed starch granules

The presence of soluble fibres in solution influenced the particle size of the

starch granules, and generally, the lower the viscosity of the medium the smaller the

size, as shown in Table 5-2. Further discussion concentrated on the volume weighted

mean (d[4,3]) will be simply referred to as particle size.

123

Table 5-2. Apparent viscosity of simulated digesta and particle size [as defined by the

volume weighted mean d(4,3)] of granules isolated from in vitro digesta before the

addition of amyloglucosidase (T0) after 1 h (T1), 2 h (T3), and 4 h (T3) of simulated

small intestinal digestion.

Treatment Apparent viscosity of digesta (Pa.s)1

T0 T1 T2 T3

Control 0.0001825d

0.0001375c

0.0001067d

0.0001193d

XG 1.515b

1.329a

0.9314a

0.7895a

GG 3.058a

1.097a

0.5548b

0.4288b

SFG 0.0762c

0.04938b

0.0423c

0.04096c

SSPS 0.02736c

0.01989b

0.01985c

0.01751c

Volume weighted mean d(4,3) (µm)

T0 T1 T2 T3

Control 18.62b 16.65

b 16.00

b 15.90

b

XG 166.9a 101.9

a 78.94

a 42.86

a

GG 175.1a 91.92

a 82.88

a 63.64

a

SFG 20.49b 18.55

b 18.33

b 18.00

b

SSPS 17.13b 16.74

b 16.73

b 16.22

b

1Apparent viscosity was measured at 50 s

-1

Figures 5-2, 5-3, 5-4, and 5-5 illustrate the particle size distribution patterns of starch

granules isolated at the beginning, following 1 h, following 2h, and at the end of

simulated small intestinal digestion, respectively. The results show primarily

unimodal particle size distributions (5 – 50 μm), which corresponds with literature

values for granule size distributions of tapioca starch, reported to be in the range of 5 -

40 μm (Eliasson, 2004). Interestingly, the granules extracted from XG and GG digesta

show bimodal size distributions patterns (5-60 and 60 – 1000 μm) and much higher

particle sizes for both (Table 5-2).

124

Figure 5-2. Particle size distribution of starch granules isolated from the in vitro

digesta of control- and fibre-fortified solutions at the beginning of simulated small

intestinal hydrolysis.

125


digesta of control- and fibre-fortified solutions after 1 hr of simulated small intestinal

hydrolysis.

126


digesta of control- and fibre-fortified solutions after 2 h of simulated small intestinal

hydrolysis.

127


digesta of control- and fibre-fortified solutions at the end of simulated small intestinal

hydrolysis.

128

Researchers working with sorghum starch have observed trimodal distributions

(Mahasukhonthachat, Sopade, and Gidley, 2010) and others working on cereal and

legume starches have reported bimodal size distributions of starch granules (White,

Doucet, Hill, and Wiseman, 2008; Kerr, Ward, McWatters, and Resurreccion, 2000).

Variations in the granular size of starch can also arise from different varieties and can

be dependent on the season during which it grows (Eliasson, 2004). Although the

source of the starch indicates the range in particle sizes, the bimodal distributions

observed for XG and GG solutions in the present study may be more related to the

effect that the gums have on granule organization and their ability to promote granule

association through aggregation (Chaisawang and Suphantharika, 2005). This will be

further evidenced and discussed in the next section.

The results demonstrate a reduction in particle size as simulated small

intestinal digestion proceeds from 0 h (Figure 5-1) to 4 h (Figure 5-4) for granules

extracted from all five treatments, indicating an increase in amylolysis of starch.

Granules isolated from the digesta of XG- and GG-fortified treatments were

significantly larger than the others. The addition of XG and GG to native tapioca

starch has been previously studied, where researchers were observing for the effects of

both gums on pasting and rheological properties of starch pastes (Chaisawang and

Suphantharika, 2006). Although they did not study particle size they were able to

demonstrate agglomeration of starch granules using microscopy and subsequently

attributed their results to the ability of XG to completely wrap and GG to form a sheet

structure around native tapioca starch granules (Chaisawang and Suphantharika,

2006). In our study, despite maintaining significantly higher particle sizes throughout

129

the simulated digestion, granules isolated from the XG and GG treatments

demonstrated a much greater degree of reductions in particle size. This may be

attributed by fewer agglomerated starch granules present (reduction in the amplitude

of the second peak) which may be caused by the dilution effect on both solutions

created by the digestive secretions of the simulated digestion model. Subsequently, the

presence of gums, which could be promoting granule association, may have been

disrupted, resulting in the decrease in particle size; however, this requires further

investigation. As discussed previously, in our earlier study we reported on the ability

of XG and GG to attenuate RSR and glucose release during simulated small intestinal

digestion; SFG and SSPS on the other hand, were only able to lower the amount of

glucose being released into the digesta, having no significant effect on RSR. This may

be partly explained by the smaller particle sizes observed for the starch granules

isolated from the digesta of both SFG and SSPS treatments.

In addition to the digestion of starch relying on the architecture and crystalline

pattern of the granule as well as non-starch components present in solution, it is also

dependent on granule size, where smaller particles are more rapidly digested due to

increased relative surface area (Choi, Woo, Ko, and Moon, 2008; Viera and Sarmento,

2008). It was previously reported that for every 100 μm increase in particle size of

maize grain the amount of starch available for amylolytic attack was reduced by 26.8

g/kg starch (Blasel, Hoffman, and Shaver, 2006). In vitro digestion of raw starch has

been observed to follow first-order kinetics (Ezeogu, Duodu and Taylor, 2005;

Wiseman, 2006) and because the granular shape of tapioca starch is round (Eliasson,

2004) the surface area is proportional to the radius of the particle. Therefore, if an

130

enzyme diffuses onto the surface of spherical particles the rate at which they are

digested is related to the average particle size (Al-Rabadi, Gilbert and Gidley, 2009;

Mahasukhonthachat et al., 2010). Although this requires further investigation,

reductions in RSR observed in our earlier study (Chapter 4) may be related to not only

the viscosity that XG and GG are able to promote but also their ability to allow

granules to come together. If the primary step in starch digestion is the diffusion of

enzymes onto the surface of starch granules then it is plausible that XG and GG are

more efficient at lowering the rate of starch hydrolysis (and therefore the release of

RSR) by not only reducing the diffusion of amylase enzymes inside the simulated

lumen but by also increasing the particle size of starch granules; subsequently, leading

to a reduction in the relative surface area being exposed to amylolytic attack.

5.3.3. Starch Morphology

5.3.3.1. Scanning electron microscopy (SEM)

SEM micrographs of native (unhydrolysed) starch granules isolated from the

five different treatments are presented in Figure 5-6. The surface of the granules

appears smooth with no indication of granular degradation, which would be evidenced

through the manifestation of cracks, pores or fissures along the exterior of the starch

granule. Although there was heterogeneity in the observed size, the majority of the

starch granules ranged between 10 and 50 µm, with some granules adhering to each

other as shown in Figure 5-6.

131

Figure 5-6. Scanning electron microscopy of tapioca starch granules isolated from

control (C)-, soy soluble polysaccharide (SSPS)-, soluble flaxseed gum (SFG)-, guar

gum (GG)-, and xanthan gum (XG)-fortified solutions.

132

Scanning electron micrographs of the granules isolated and purified from GG- and

XG-fortified solutions displayed pockets of larger agglomerated starch particles, as

shown in Figure 5-7.

Figure 5-7. Scanning electron micrographs of tapioca starch aggregates isolated from

guar gum (GG)- and (XG)-fortified solutions.

133

Our earlier analyses of particle size of isolated starch granules revealed bimodal

distributions for these two treatments where the second peak indicated sizes > 100

µm, which is likely due to the observable aggregation of starch granules in the

presence of GG and XG. In our earlier study we reported on the ability of both GG

and XG to form a gel network in the presence of starch and it is plausible that both

gums are able to enwrap the starch granules thereby promoting aggregation inside the

gel matrix. This observation was also made in two earlier studies exploring the pasting

properties of tapioca starch pastes with the addition of GG and XG (Chaisawang and

Suphantharika, 2005; Chaisawang and Suphantharika, 2006). Moreover, Gonera and

Cornillon (2002) reported on the ability of XG to adhere to the surface of starch

granules inducing a stabilizing effect. In our study, despite observed differences in the

ability of soluble fibres to promote starch aggregation, the surface topography of

isolated granules was observed to be smooth with no evidence of starch degradation in

either of the five treatment types.

Starch granules isolated from the in vitro digesta at the beginning of simulated

small intestinal digestion (third stage of in vitro digestion) of the control and fibre-

fortified treatments exhibited variability in their morphological characteristics, as

shown in Figure 5-8.

134

Figure 5-8. Morphology of granules isolated from the in vitro digesta of the control

and fibre foritifed treatments at the beginning of simulated small intestinal digestion.

135

Preincubation with simulated salivary (5 min), gastric (1 h) and small

intestinal fluids (0 h, before the addition of amyloglucosidase) showed granules with

pitting along the exterior, which is the first step in the amylolysis of starch (Eliasson,

2004). Moreover, the granules extracted from the in vitro digesta of the control, SSPS

and SFG treatments showed further evidence of structural degradation through the

appearance of exo-corrosion (surface erosion). Changes in surface topography through

both pitting and exo-corrosion were less apparent in the GG and XG treatments, with

only the former exhibiting pitting along the surface exterior. Since there is no

dissimilarity in the structure of starch (amylose/amylopectin ratio, amylopectin branch

density, and polymorphic form) and there was no heat-moisture treatment during

formulation, which has been shown to promote starch hydrolysis (Varatharajan,

Hoover, Li, Vasanthan, Nantanga, Seetharaman, Liu, Donner, Jaiswal, and Chibbar,

2011) the variation in the degree of hydrolysis is likely due to the presence of fibres

and ensuing difference in the viscosity of the surrounding medium. In our study, as

simulated small intestinal digestion continued changes in starch morphology became

more apparent, as shown in Figures 5-9 and 5-10.

136


and fibre fortified treatments after 1 h of simulated small intestinal digestion.

137


and fibre fortified treatments after 2 h of simulated small intestinal digestion.

138

Both exo- and endocorrosion were observed for starch granules isolated from the

digesta of the control, SSPS and SFG treatments. As was discussed, the first alteration

in the structure of starch granules during digestion is pitting at the surface (Eliasson,

2004). Moreover, it has been shown that these truncatures are weak points in starch

granules that subsequently increase susceptibility to further hydrolysis (Chen, Huang,

Tang, Chen and Zhang, 2011; Franco, Cabral, Tavares, 2002). Pits become enlarged,

as seen in the above micrograph images, and form numerous pores or channels further

weakening the structure (Dona et al., 2010). Subsequently amylase enzymes enter

these channels to gain access into the granule interior and digestion proceeds as the

enzymes preferentially hydrolyse the less organized (amorphous) region surrounding

the hilum. Some pitting and surface degradation became apparent after 1 h simulated

small intestinal digestion for the granules isolated from the GG-fortified digesta and

only after 2 h were any changes in surface topography observed for the granules

isolated from the XG digesta.

139


and fibre fortified treatments at the end of in vitro digestion.

140

Starch morphology after 4 h in vitro digestion of the control treatment shows

much larger pores with exposed interior segments as well as broken granule fragments

surrounding the granules. These changes are also observed for SSPS, SFG and, to a

lesser extent, GG treatments indicating a greater degree of amylolysis of starch in

comparison to granules isolated from the XG digesta. Although the micrographs

display images of isolated/purified starch granules, it is clear that the changes

occurring to starch morphology are inherently dependent on the medium from which

they were extracted. As shown earlier, both GG and XG demonstrated the greatest

ability to resist reductions in viscosity, with the latter retaining a considerable amount

of its viscosity even after being exposed to the secretions of the in vitro digestion

model. The SEM micrographs show noticeable variability in the hydrolysis of starch

through the inclusion of different fibres. In our earlier study we observed the ability of

XG to attenuate glucose diffusion across a dialysis membrane and reported on the

inverse relationship that existed between digesta viscosity and glucose diffusivity.

Although we attributed this to the ability of fibres to provide a physical barrier to the

permeability of glucose into the dialysate we speculated on potential effects on the

hydrolysis of starch inside the simulated digesta. This study shows evidence that there

is a discernible difference in the ability of soluble fibres to lower the rate at which

starch granules are digested, where those that are better able to resist alterations in

rheological behaviour during in vitro digestion are more effective in moderating

digestion rates.

141

5.3.3.2. Light microscopy of digested starch granules

The effect of simulated small intestinal digestion on populations of starch

granules are shown in Figure 5-12.

Figure 5-12. Light microscopy of starch isolated from the digesta of control (C), soy

soluble polysaccharide (SSPS), soluble flaxseed gum (SFG), guar gum (GG) and

xanthan gum (XG) treatments at the beginning (T1), after 1 h (T2), after 2 h (T3) and

at the end of simulated small intestinal digestion (T4).

142

In conjunction with the SEM micrographs, continued digestion led to pronounced

changes in starch morphology, as evidenced by the manifestation of channels and

openings along the granule exterior. These changes are present at the initial stage of

digestion (T0 images) for the control, SSPS, and SFG-isolated granules, indicating a

greater susceptibility to amylolysis in comparison to starch granules extracted from

the digesta of GG and XG treatments. Upon completion of in vitro digestion, some

granules were completely hydrolysed while others showed evidence of fragmentation

with residual starch being left behind. Interestingly, light microscopy revealed that at

the end of simulated digestion many granules remained intact, even for the control

treatment, demonstrating heterogeneity in starch hydrolysis during in vitro digestion.

Hasjim et al. (2010) made a similar observation where they reported on the presence

of undigested granules even after 25 h of in vitro digestion. This finding might explain

why in our earlier study we reported on only 40% of the starch being hydrolysed for

the control treatment. Moreover, in that study we measured starch digestibility by

computing the rate of RSR and observed that XG had the most pronounced effect on

lowering the rate of release of sugars, followed by GG. Although we attributed this to

an increase in digesta viscosity, it remained unclear whether the result was due to the

effect the soluble fibres exerted on the hydrolysis of starch or whether there was

another mechanism by which they were modulating the release of sugars. The light

microscopy images confirm the variability in the digestion pattern of tapioca starch

and the influence of soluble fibres.

Although our present study does not look at the effects of soluble fibre

inclusion on enzyme activity, it is important to discuss the potential effects they may

143

have on moderating such events. It has been reported that if amylase activity can be

inhibited by controlling parameters that affect amylase diffusion then there could be

nutritional implications (Al-Rabadi, et al., 2009), such as those related to reductions in

postprandial glycemia. As was mentioned in our previous discussion regarding starch

particle size, the digestion of both cooked and raw starches follows first order kinetics

(Equation 5-1) (Goni, Garcia-Alonso and Saura-Calixto, 1997)

𝐶 = 𝐶∞(1 − 𝑒−𝑘𝑡) (Equation 5-1)

where C is the concentration of glucose at a given time, C∞ is the concentration of

glucose upon completion of the reaction and k is a first order kinetic constant.

However, starch hydrolysis is a complex system where many factors have the

potential of influencing the rate of the reaction and subsequent glucose release, such

as by means of inhibiting enzyme activity. Inhibition of enzyme activity can either be

competitive or non-competitive.

During competitive-inhibition the inhibitor directly binds to the enzyme

thereby preventing the binding of the substrate to the enzymes active site.

Subsequently, a variety of models exist that can be used to explain the enzyme

kinetics of starch digestibility, with some integrating but slightly deviating from the

well-known Michaelis-Menten model (Fujii and Kawamura, 1985; Wang, Zeng, Liu

and Yuan, 2006; Dona et al., 2010) (Equation 5-2).

𝐷[𝑃]

𝐷𝑡=

𝑉𝑚𝑎𝑥[𝑆]

𝐾𝑚 (1+ [𝑃]

𝐾𝑖)+ [𝑆]

(Equation 5-2)

144

where [P] is the concentration of product at any given time, Vmax is the maximum

velocity of the reaction, Km is the Michaelis constant (from the Michaelis-Menten

equation), [S] is the substrate concentration and Ki is the product inhibition constant.

The proposed model has been used to describe the product inhibition of glucoamylase

and amylase by glucose. However, there is a so-called “grace zone” where inhibition

is absent below a threshold concentration of glucose (Wang et al., 2006). Moreover,

many studies modelling product-inhibited reactions lack reasonable explanations as to

why the proposed time courses show deviation from the shape of Michaelis-Menten

reactions and therefore it has been reported that product inhibition reactions are not

useful in modelling starch digestion curves (Dona et al., 2010). During non-

competitive inhibition, the inhibitor has an indistinguishable affinity for the enzyme

and uncompetitive inhibition is where the inhibitor binds solely with the enzyme-

substrate complex (Kuchel and Ralston, 1997); in either case the result is a decrease in

the rate of the reaction. Although both types of inhibition (competitive or non-

competitive/uncompetitive) may occur during starch digestion, physical properties of

the digesta, such as viscosity, may also influence the rate of starch hydrolysis by

inducing diffusional restrictions (Dona et al., 2010).

As we have shown in our study, the presence of soluble fibres in solution has a

tremendous impact on their subsequent rheological properties. Moreover, SEM and

light microscopy images indicate the plausible effect that fibres exert on the

amylolysis of starch granules, likely due to the increase in digesta viscosity.

Therefore, it can be assumed that kinetic behaviour can be partly influenced by

diffusional restrictions, as shown by equation 5-3, which demonstrates reduced rates

145

of reaction by increasing substrate concentration (Pastrana, Gonzalez, Miron and

Murado, 1998).

𝐷[𝑃]

𝐷𝑡=

𝑉𝑚𝑎𝑥[𝑆]

𝐾𝑚 + [𝑆] 𝑒𝐷𝑣𝑅 (Equation 5-3)

where Dv is a first order coefficient used to characterize the relationship between the

rate of the reaction and diffusional restrictions and the term ‘R’ is derived from:

𝑅 = 𝑅𝑚 + (𝑅0 − 𝑅𝑚)𝑒−𝜇[𝑆] (Equation 5-4)

where Ro is the diffusional restriction when [S]=0 and Rm is the largest value

attainable for the diffusional restriction and μ is a diffusional restriction coefficient.

Although the equation assumes increasing substrate concentration translates into

higher viscosities and therefore greater diffusion restrictions, it does not account for

variability amongst different polysaccharides. In our system, much higher

concentrations were required for SSPS and SFG in order to match the apparent

viscosity of XG and GG. Moreover, the viscosity inside the simulated digesta was

much higher for XG, which also had a lower initial concentration. Therefore, viscosity

may be a more valid mechanistic indicator of diffusional restrictions than

concentration. This is especially true as the addition of XG to starch solutions led to

the most pronounced reductions in starch hydrolysis as evidenced by the changes in

starch morphology as well as the effects we observed on RSR and glucose diffusivity

(Chapter 4). Nonetheless, a mathematical model incorporating a diffusion coefficient

may be useful in attempting to relate soluble fibre functionality to starch hydrolysis.

146

The quantitative analysis of starch digestion and mathematical and empirical

modeling of enzyme kinetics can be very useful tools in advancing our understanding

of starch digestibility. Moreover, they may help researchers better understand how

complex polymer systems influence the rate at which starch is hydrolysed and how the

presence of soluble fibres might play a role at promoting glycemic reductions.

5.4.Conclusion

Starch suspensions were formulated with native tapioca starch and different

soluble fibres. Formulations were designed to produce equi-viscous solutions at lower

shear rates. Subsequently, starch granules were isolated and purified from the digesta

at various times of simulated small intestinal digestion for analysis of particle size and

starch morphology. Granules extracted from the digesta of the control, SSPS and SFG

treatments showed unimodal size distributions (5 – 50 μm) whereas those purified

from the GG and XG treatments showed bimodal distributions (5 – 60 μm, 60 – 1000

μm), which we believe is attributed to the ability of both GG and XG to enwrap starch

granules allowing for granular associations, as evidenced by the micrograph images.

The size of starch granules decreased as digestion proceeded due to the continued

hydrolysis of starch. The SEM micrographs revealed pitting and surface erosion at the

beginning of simulated small intestinal digestion for granules extracted from low

viscosity digesta (control, SSPS, SFG). As digestion continued these changes became

more apparent as both exo- and endo-corrosion were visible showing extensive

degradation of starch granules, which was also observed for the GG treatment.

Granules extracted from the digesta of XG-fortified solutions showed fewer changes

in surface topography, which we believe is due to the ability of XG to retain a

147

measurable amount of its viscosity throughout in vitro digestion. Light microscopy

showed that despite variability in the extent of starch hydrolysis between the different

treatments there was also heterogeneity in the hydrolysis of starch within each

treatment (some granules remained unaltered while others were completely degraded).

The results of our previous studies, where we explored the ability of soluble

fibres to attenuate RSR, glucose liberation and diffusion, may be explained by their

ability, or inability, to affect the hydrolysis of starch. Although the agreement amongst

researchers studying the glucose-lowering ability of soluble fibres is that the effects

are due to the increase in digesta viscosity; the mechanism(s) by which fibres might

modulate the in vitro glycemic response remain unclear. From our study, it is apparent

that fibre-fortification leads to attenuations in the hydrolysis of starch. Moreover, the

effects are only observed for fibres that are able to retain a measurable amount of

viscosity during simulated digestion, where those less viscous (during digestion) are

much less effective. It is conceivable then that increasing the viscosity of the

surrounding medium leads to less effective amylolysis of starch granules through

interfering with the diffusion of enzymes onto the surface of starch granules. The

effects may also be due to the ability of some gums to allow starch granules to

agglomerate thereby reducing the relative surface area exposed to digestive enzymes.

To augment the findings of this study, it would be vital to explore the effects that

soluble fibres might have on enzyme (amylase, glucoamylase) activity to gain further

understanding of the mechanism by which soluble fibres are able to modulate the in

vitro glycemic response.

148

6. Study strengths and limitations

There are a number of strengths associated with this work. Investigations focused

on understanding the functionality of soluble dietary fibres as they relate to glycemic

reductions often employ clinical trials to observe the effects of fibre fortification.

Subsequently, the results of many of these studies show positive outcomes (enhanced

glycemic reductions) and the discussion is centred on the increase in viscosity that

fibres are able to generate. Although such studies are fundamentally imperative they

pose many questions concerning the mechanism of action. As outlined in Chapter 2,

the GIT is a complex system and the breakdown of starch into glucose occurs in

several stages inside the small intestine. Many in vitro and in vivo studies use a single

marker, most commonly glucose levels, as a measure of functionality from which a

discussion and theoretical models are generated. Our experimental setup included

looking at the different stages of starch hydrolysis, attempting to observe how fibre-

fortification may lead to reductions in RSR (simulated luminal digestion), glucose

release (simulated membrane-level digestion) and finally glucose diffusion.

Understanding how one or more of these events are influenced by dietary fibres will

help generate a more balanced and comprehensive understanding of their glucose-

lowering ability.

Despite possessing physicochemical characteristics that allow fibres to thicken

when dispersed in solution, the secretions of the GIT have a profound influence on

their rheological properties, which as a result may alter their in vivo functionality. In

our study, we closely monitor changes in rheological behaviour (both viscosity and

viscoelastic properties) of different dietary fibres in order to gain a snapshot of their

149

“physiological flow behaviour”, which is one that should be looked at when

attempting to relate viscosity or gelation to their purported physiological effects, in

this case glycemic reductions.

Moreover, this study uses different techniques to attain both quantitative and

qualitative information to divulge the functionality of dietary fibres using the

prescribed in vitro system. By extracting whole starch granules from the simulated

digesta we were able to visualize the changes occurring when starch was dispersed

both in the absence and presence of select dietary fibres. Light scattering data and

micrograph images augment the RSR and glucose release data we acquired, which

create a more balanced approach in studying the ability of soluble fibres to moderate

the in vitro glycemic response.

The limitations of the study are present just like in any other in vitro digestion

model where mimicking the in vivo conditions can never be fully accurate. Despite

using a three-stage digestion model where we were able to control many essential

elements, such as temperature, pH, agitation/mixing and enzyme activity, these events

are highly dynamic and influenced by a variety of factors, as we outlined in Chapter 2.

Neurohumoral regulation, gastric emptying and muscular contractions, variability in

shear rates and the rate of release and levels of secretions are difficult to mimic using

a static digestion model. Nonetheless, using such models allows researchers to gain a

better understanding of how food materials and polymer systems function inside the

GIT, despite the accompanying limitations in the design protocol.

150

7. Conclusion and future directions

The ability of different dietary fibres to influence the in vitro glycemic response

was the focus of the present study. Xanthan gum (XG), guar gum (GG), soluble

flaxseed gum (SFG) and soy soluble polysaccharide (SSPS) were added to protein-

starch solutions. Subsequently, all solutions were subjected to a three-stage in vitro

digestion model mimicking the salivary, gastric and small intestinal steps, and their

rheological behaviour was monitored throughout. The concentrations of fibres were

selected to create similar apparent viscosities at lower shear rates (20 – 50 s-1

), to

mimic the shear rates reported during digestion (Borwankar, 1992; Steffe, 1996). Prior

to simulated digestion, all solutions exhibited pseudoplastic flow as indicated by their

behaviour index values. Moreover, oscillatory testing revealed the presence of

viscoelastic gels for all solutions; however, rheological behaviour was greatly

influenced by the secretions of the digestion model.

Despite showing comparable viscosities in solution, the flow behaviours of the

different treatments were drastically changed as they progressed through each stage of

simulated digestion. The decrease in viscosity was in the order of: control < SSPS <

SFG < GG < XG where it was observed that only GG and XG retained a measurable

amount of their initial viscosity. The control, SSPS and SFG behaved as Newtonian

fluids by the time in vitro digestion had culminated. Moreover, of the four fibre-

fortified treatments, XG was the only one able to retain its solid-like network, while

GG behaved as a concentrated solution and SSPS and SFG both displayed complete

loss of G’ (large value for tan δ). Interestingly, the loss in viscosity throughout

simulated small intestinal digestion was nearly instantaneous for the control, SSPS

151

and SFG treatments whereas the losses were observed to be more gradual in the GG

and XG treatments. We attributed this to the variability in the hydrolysis of the starch

component, where both GG and XG are likely influencing the rate of digestion of

starch, therefore allowing for a steadier reduction in flow behaviour.

The ability of soluble fibres to moderate starch hydrolysis was examined by

measuring the amount of reducing sugars released inside the simulated digesta. There

was an inverse relationship between digesta viscosity and reducing sugar release

(RSR) where the most viscous treatment (XG) elicited the greatest reduction in both

the final concentration and the rate of release. As the viscosity of the digestive

contents increased it is plausible that the mixing of the secretions with the solutions

becomes less effective, allowing for a steadier hydrolysis of starch due to a decrease

in amylolytic attack. Despite having a negligible effect on RSR, when we measured

the concentration of glucose in the digesta even the treatments with much lower

viscosities (SFG and SSPS) were able to moderate the amount of glucose being

generated inside the digesta in comparison to the control. The apparent viscosities of

the fibre-fortified digest of SSPS and SFG ranged between 60 and120 mPa.s,

respectively (higher than what was observed for the control treatment, which had

values similar to that of water). When simulated digestion was carried out inside a

dialysis system only XG (and to a lesser extent, GG), which had measurable elastic

components throughout digestion, were able to attenuate the rate of diffusion of

glucose. The results indicate that the functionality of soluble fibres, as related to their

ability to attenuate the in vitro glycemic response, may be multi-faceted and reliant on

more than just their ability to increase viscosity. The capacity of fibres to maintain a

152

gel network inside the simulated lumen may be crucial in allowing them to entrap

glucose molecules thereby lowering their rate of diffusion into the dialysate.

In the final portion of the study, replacing pregelatinized starch with native

tapioca starch allowed for more accurate investigation of starch degradation by

observing changes in particle size and granule morphology. Particle size distribution,

as defined by the volume weighted mean - d[4,3] value, showed unimodal distribution

(5 to 50 μm) for isolated starch granules. The addition of XG and GG to starch

suspensions resulted in a shift to bimodal size distributions demonstrating an

aggregating effect on starch granules, which was confirmed by both SEM and light

microscopy (granule bridging led to aggregates > 100 μm). Light scattering showed a

drop in particle size throughout simulated small intestinal digestion due to the

amylolysis of starch granules. Micrograph images of starch showed spherical granules

with smooth surfaces and no evidence of structural degradation caused by the

isolation/purification method employed. Extracting granules from the in vitro digest

showed heterogeneity in changes done to surface topography where those extracted

from less viscous system (control, SFG and SSPS) showed evidence of continued and

extensive hydrolysis as time went on. Pitting along the surface, followed by the

formation of crack (exo-corrosion) were the first changes observed. Subsequently,

extensive degradation of the granule interior (endo-corrosion) was observed with

some granules having been fully digested upon completion of in vitro digestion. The

progression of these changes was delayed in granules isolated from more viscous

systems and only after 4 h small intestinal digestion of the XG-fortified treatment did

granules show signs of amylolysis. These findings augment the earlier portion of the

153

study and highlight the ability of viscous fibre to attenuate the hydrolysis of starch.

This study shows that fibre-fortification may have variable effects on the hydrolysis of

starch and subsequent release and diffusion of glucose. Despite the substantiation of

our study of the notion that viscous fibres are able to modulate glucose levels, other

investigations would be essential in allowing us to fully understand the mechanism by

which they are able to attenuate blood glucose levels.

In tandem with the findings of our study, exploring the potential (inhibitory)

effects of dietary fibres on enzyme activity will be essential in allowing researchers to

better understand the mechanism surrounding their glucose-lowering ability. Studies

have shown that α-amylase activity can be inhibited by polyphenolic compounds that

are present in high-fibre foods (Sreerama, Sashikala, and Pratape, 2012). Moreover,

results of several studies have reported that anthocyanins found in fibre-rich foods

such as fruits and vegetables have both in vivo and in vitro inhibitory activity towards

α-glucosidase (McDougall, Dobson, Shpiro, Smith, Stewart, and Fyffe, 2005) and that

the GI values of foods were shown to be inversely related to their respective

polyphenol content (Ramdath, Padhi, Hawke, Sivaramalingam, and Tsao, 2014).

Although the studies focus on anthocyanin content and that the reports are

contradictory, with some studies showing that some compounds such as lignin may in

fact stimulate enzyme activity (Zhang, Cui, Yin, Sun, and Li, 2013), it would be

essential to explore how soluble fibres, which have been observed to attenuate starch

hydrolysis, might play a role in mediating enzyme activity. Although antidiabetic

drugs, such as acarbose and miglitol, which are designed to inhibit amylase activity

are available on the market they are associated with a range of undesirable side effects

154

(Etxeberria, de La Garza, Campion, Martinez, and Milagro, 2012; Lordan, Smyth,

Soler-Vila, Stanton, and Ross, 2013); it would therefore be more beneficial to

discover natural sources to augment current strategies in diabetic therapies. The

digestibility and glucose-release behaviour of starch is of both metabolic and clinical

importance. As a result, future investigations on soluble fibre structure-function

relationships are essential to researchers seeking to develop novel functional foods

designed to promote glycemic reductions.

155

References

Adeghate, E., Schattner, P., & Dunn, E. (2006). An update on the etiology and

epidemiology of diabetes mellitus. Annals of the New York Academy of Sciences,

1084(1), 1-29.

Al-Rabadi, G. J., Gilbert, R. G., & Gidley, M. J. (2009). Effect of particle size on

kinetics of starch digestion in milled barley and sorghum grains by porcine alpha-

amylase. Journal of Cereal Science, 50(2), 198-204.

Alzueta, C., Rodriguez, M., Cutuli, M., Rebole, A., Ortiz, L., Centeno, C., et al.

(2003). Effect of whole and demucilaged linseed in broiler chicken diets on

digesta viscosity, nutrient utilisation and intestinal microflora. British Poultry

Science, 44(1), 67-74.

Anderson, J. W., Randles, K. M., Kendall, C. W., & Jenkins, D. J. (2004).

Carbohydrate and fiber recommendations for individuals with diabetes: A

quantitative assessment and meta-analysis of the evidence. Journal of the

American College of Nutrition, 23(1), 5-17.

Anderson, G. H., Cho, C. E., Akhavan, T., Mollard, R. C., Luhovyy, B. L., &

Finocchiaro, E. T. (2010). Relation between estimates of cornstarch digestibility

by the englyst in vitro method and glycemic response, subjective appetite, and

short-term food intake in young men. The American Journal of Clinical

Nutrition, 91(4), 932-939.

156

ANZFA (Australia New Zealand Food Authority) (2000) Notice of a Proposed

Change to Food Regulation and Further Invitation for Submissions. Application

277 . Inulin and Fructooligosaccharides as Dietary Fibre.

Aravind, N., Sissons, M., Egan, N., & Fellows, C. (2012). Effect of insoluble dietary

fibre addition on technological, sensory, and structural properties of durum wheat

spaghetti. Food Chemistry, 130(2), 299-309.

Au, M. M., Goff, H. D., Kisch, J. A., Coulson, A., & Wright, A. J. (2013). Effects of

soy-soluble fiber and flaxseed gum on the glycemic and insulinemic responses to

glucose solutions and dairy products in healthy adult males. Journal of the


Barclay, A. W., Petocz, P., McMillan-Price, J., Flood, V. M., Prvan, T., Mitchell, P.,

et al. (2008). Glycemic index, glycemic load, and chronic disease risk--a meta-

analysis of observational studies. The American Journal of Clinical Nutrition,

87(3), 627-637.

Behall, K. M., Scholfield, D. J., & Hallfrisch, J. (2005). Comparison of hormone and

glucose responses of overweight women to barley and oats. Journal of the


Blasel, H., Hoffman, P., & Shaver, R. (2006). Degree of starch access: An enzymatic

method to determine starch degradation potential of corn grain and corn silage.

Animal Feed Science and Technology, 128(1), 96-107.

http://www.foodstandards.gov.au/_srcfiles/A277_FA.pdf

http://www.foodstandards.gov.au/_srcfiles/A277_FA.pdf

157

Bobboi, A., & Stephens, A. (1996). The effects of electrolyte and hydrogen ion

concentrations on guar gum and glucose tolerance following intraduodenal

administration. Nutrition Research, 16(8), 1403-1409.

Boron, W. F., & Boulpaep, E. L. (2009). Medical physiology. A cellular and

molecular approach. saunders.

Borwankar, R. P. (1992). Food texture and rheology: A tutorial review. Journal of

Food Engineering, 16(1), 1-16.

Braaten, J. T., Wood, P. J., Scott, F. W., Riedel, K. D., Poste, L. M., & Collins, M. W.

(1991). Oat gum lowers glucose and insulin after an oral glucose load. The

American Journal of Clinical Nutrition, 53(6), 1425-1430.

Brenelli, S., Campos, S., & Saad, M. (1997). Viscosity of gums in vitro and their

ability to reduce postprandial hyperglycemia in normal subjects. Brazilian

Journal of Medical and Biological Research, 30(12), 1437-1440.

Brennan, C. S. (2005). Dietary fibre, glycaemic response, and diabetes. Molecular

Nutrition & Food Research, 49(6), 560-570.

Brennan, C. S., & Cleary, L. J. (2005). The potential use of cereal (1→ 3, 1→ 4)-β-D-

glucans as functional food ingredients. Journal of Cereal Science, 42(1), 1-13.

Brennan, C. S., & Tudorica, C. M. (2008). Evaluation of potential mechanisms by

which dietary fibre additions reduce the predicted glycaemic index of fresh

pastas. International Journal of Food Science & Technology, 43(12), 2151-2162.

158

Brennan, C., Blake, D., Ellis, P., & Schofield, J. (1996). Effects of guar

galactomannan on wheat bread microstructure and on theIn vitroandIn

vivoDigestibility of starch in bread. Journal of Cereal Science, 24(2), 151-160.

Brennan, M. A., Derbyshire, E. J., Brennan, C. S., & Tiwari, B. K. (2012). Impact of

dietary fibre‐enriched ready‐to‐eat extruded snacks on the postprandial glycaemic

response of non‐diabetic patients. Molecular Nutrition & Food Research, 56(5),

834-837.

Brennan, M. A., Monro, J. A., & Brennan, C. S. (2008). Effect of inclusion of soluble

and insoluble fibres into extruded breakfast cereal products made with reverse

screw configuration. International Journal of Food Science & Technology,

43(12), 2278-2288.

Brighenti, F., Pellegrini, N., Casiraghi, M., & Testolin, G. (1995). In vitro studies to

predict physiological effects of dietary fibre. European Journal of Clinical

Nutrition, 49(S12), S81-S88.

Brown, K. H., Perez, F., Peerson, J. M., Fadel, J., Brunsgaard, G., Ostrom, K. M., et

al. (1993). Effect of dietary fiber (soy polysaccharide) on the severity, duration,

and nutritional outcome of acute, watery diarrhea in children. Pediatrics, 92(2),

241-247.

Brownlee, I. A. (2011). The physiological roles of dietary fibre. Food Hydrocolloids,

25(2), 238-250.

159

Brownlee, I. A., Dettmar, P. W., Strugala, V., & Pearson, J. P. (2006). The interaction

of dietary fibres with the colon. Current Nutrition & Food Science, 2(3), 243-

264.

Burkitt, D. P., & Trowell, H. C. (1977). Dietary fibre and western diseases. Irish

Medical Journal, 70(9), 272-277.

Byrnes, S. E., Miller, J. C., & Denyer, G. S. (1995). Amylopectin starch promotes the

development of insulin resistance in rats. The Journal of Nutrition, 125(6), 1430-

1437.

Cairns, P., Miles, M., & Morris, V. (1986). Intermolecular binding of xanthan gum

and carob gum.

Cameron-Smith, D., Collier, G., & O'dea, K. (1994). Effect of soluble dietary fibre on

the viscosity of gastrointestinal contents and the acute glycaemic response in the

rat. British Journal of Nutrition, 71(04), 563-571.

Camilleri, M., Malagelada, J. R., Brown, M. L., Becker, G., & Zinsmeister, A. R.

(1985). Relation between antral motility and gastric emptying of solids and

liquids in humans. The American Journal of Physiology, 249(5 Pt 1), G580-5.

Canadian Diabetes Association. (2009). The prevalence and costs of diabetes. clinical

practice guidelines. Retrieved September 17, 2014, from

http://www.diabetes.ca/publications-newsletters/advocacy-reports/economic-

tsunami-the-cost-of-diabetes-in-canada

http://www.diabetes.ca/publications-newsletters/advocacy-reports/economic-tsunami-the-cost-of-diabetes-in-canada

http://www.diabetes.ca/publications-newsletters/advocacy-reports/economic-tsunami-the-cost-of-diabetes-in-canada

160

Chaisawang, M., & Suphantharika, M. (2005). Effects of guar gum and xanthan gum

additions on physical and rheological properties of cationic tapioca starch.

Carbohydrate Polymers, 61(3), 288-295.

Chaisawang, M., & Suphantharika, M. (2006). Pasting and rheological properties of

native and anionic tapioca starches as modified by guar gum and xanthan gum.

Food Hydrocolloids, 20(5), 641-649.

Chang, E. B., Sitrin, M. D., & Black, D. D. (1996). Gastrointestinal, hepatobiliary,

and nutritional physiology. JB Lippincott Company.

Chawla, R., & Patil, G. (2010). Soluble dietary fiber. Comprehensive Reviews in Food

Science and Food Safety, 9(2), 178-196.

Chen, Y., Huang, S., Tang, Z., Chen, X., & Zhang, Z. (2011). Structural changes of

cassava starch granules hydrolyzed by a mixture of α-amylase and glucoamylase.


Chessa, S., Huatan, H., Levina, M., Mehta, R., Ferrizzi, D., & Rajabi-Siahboomi, A.

(2014). Application of the dynamic gastric model to evaluate the effect of food on

the drug release characteristics of a hydrophilic matrix formulation. International

Journal of Pharmaceutics, 466(1), 359-367.

Chivero, P., Gohtani, S., Ikeda, S., & Nakamura, A. (2014). The structure of soy

soluble polysaccharide in aqueous solution. Food Hydrocolloids, 35, 279-286.

161

Codex Alimentarius Commission, ALINORM 10/33/26, Report of the 31st Session of the

Codex Committee on Nutrition and Foods for Special Dietary Uses, Düsseldorf,

Germany, 2 - 6 November 2009 Retrieved September 11, 2014, from

http://ftp.fao.org/codex/Alinorm10/al33_26e.pdf

Cui, W., & Mazza, G. (1996). Physicochemical characteristics of flaxseed gum. Food

Research International, 29(3), 397-402.

Cunnane, S., & Thompson, L. U. (1995). Flaxseed in human nutrition. AOCS Press.

Dhital, S., Warren, F. J., Zhang, B., & Gidley, M. J. (2014). Amylase binding to starch

granules under hydrolysing and non-hydrolysing conditions. Carbohydrate

Polymers, 113, 97-107.

Dikeman, C. L., & Fahey Jr, G. C. (2006). Viscosity as related to dietary fiber: A

review. Critical Reviews in Food Science and Nutrition, 46(8), 649-663.

Dikeman, C. L., Murphy, M. R., & Fahey, G. C.,Jr. (2006). Dietary fibers affect

viscosity of solutions and simulated human gastric and small intestinal digesta.

The Journal of Nutrition, 136(4), 913-919.

Ding, H. H., Cui, S. W., Goff, H. D., Wang, Q., Chen, J., & Han, N. F. (2014).

Soluble polysaccharides from flaxseed kernel as a new source of dietary fibres:

Extraction and physicochemical characterization. Food Research International,

56, 166-173.

http://ftp.fao.org/codex/Alinorm10/al33_26e.pdf

162

Dona, A. C., Pages, G., Gilbert, R. G., & Kuchel, P. W. (2010). Digestion of starch: In

vivo and in vitro kinetic models used to characterise oligosaccharide or glucose

release. Carbohydrate Polymers, 80(3), 599-617.

Doublier, J., & Cuvelier, G. (1996). Gums and HydrocoIIoids functional aspects.

Carbohydrates in Food, (74), 283.

Eastwood, M. A., & Morris, E. R. (1992). Physical properties of dietary fiber that

influence physiological function: A model for polymers along the gastrointestinal

tract. The American Journal of Clinical Nutrition, 55(2), 436-442.

Edwards, C. H., Warren, F. J., Milligan, P. J., Butterworth, P. J., & Ellis, P. R. (2014).

A novel method for classifying starch digestion by modelling the amylolysis of

plant foods using first-order enzyme kinetic principles. Food & Function, 5(11),

2751-2758.

Edwards, C. A., Johnson, I. T., & Read, N. W. (1988). Do viscous polysaccharides

slow absorption by inhibiting diffusion or convection? European Journal of

Clinical Nutrition, 42(4), 307-312.

EFSA (European Food Safety Authority). Outcome of the Public consultation on the

Draft Opinion of the Scientific Panel on Dietetic products, Nutrition, and

Allergies (NDA) on Dietary Reference Values for carbohydrates and dietary

fibre. EFSA Journal 2010; 8(5):1508.

Eliasson, A. (2004). Starch in food: Structure, function and applications CRC Press.

163

Ellegård, L. H., Andersson, S. W., Normén, A. L., & Andersson, H. A. (2007).

Dietary plant sterols and cholesterol metabolism. Nutrition Reviews, 65(1), 39-45.

Ellegård, L., & Andersson, H. (2007). Oat bran rapidly increases bile acid excretion

and bile acid synthesis: An ileostomy study. European Journal of Clinical

Nutrition, 61(8), 938-945.

Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classification and

measurement of nutritionally important starch fractions. European Journal of

Clinical Nutrition, 46 Suppl 2, S33-50.

Englyst, K. N., Englyst, H. N., Hudson, G. J., Cole, T. J., & Cummings, J. H. (1999).

Rapidly available glucose in foods: An in vitro measurement that reflects the

glycemic response. The American Journal of Clinical Nutrition, 69(3), 448-454.

Etxeberria, U., de la Garza, A. L., Campión, J., Martínez, J. A., & Milagro, F. I.

(2012). Antidiabetic effects of natural plant extracts via inhibition of

carbohydrate hydrolysis enzymes with emphasis on pancreatic alpha amylase.

Expert opinion on therapeutic targets, 16(3), 269-297.

Ezeogu, L., Duodu, K., & Taylor, J. (2005). Effects of endosperm texture and cooking

conditions on the in vitro starch digestibility of sorghum and maize flours.

Journal of Cereal Science, 42(1), 33-44.

Fabek, H. (2011). Effect of in Vitro Human Digestion on the Viscosity of

Hydrocolloids in Solution: A Dietary Fibre Study (Master’s dissertation),

164

Fabek, H., Messerschmidt, S., Brulport, V., & Goff, H. D. (2014). The effect of in

vitro digestive processes on the viscosity of dietary fibres and their influence on

glucose diffusion. Food Hydrocolloids, 35, 718-726.

Fafaungwithayakul, N., Hongsprabhas, P., & Hongsprabhas, P. (2011). Effect of soy

soluble polysaccharide on the stability of soy-stabilised emulsions during in vitro

protein digestion. Food Biophysics, 6(3), 407-415.

Ferrua, M. J., & Singh, R. P. (2010). Modeling the fluid dynamics in a human

stomach to gain insight of food digestion. Journal of Food Science, 75(7), R151-

62.

Field, C. J., McBurney, M. I., Massimino, S., Hayek, M. G., & Sunvold, G. D. (1999).

The fermentable fiber content of the diet alters the function and composition of

canine gut associated lymphoid tissue. Veterinary Immunology and

Immunopathology, 72(3), 325-341.

Fodje, A. M., Chang, P. R., & Leterme, P. (2009). In vitro bile acid binding and short-

chain fatty acid profile of flax fiber and ethanol co-products. Journal of

Medicinal Food, 12(5), 1065-1073.

Franco, C. M., Cabral, R. A., & Tavares, D. Q. (2002). Structural and

physicochemical characteristics of lintnerized native and sour cassava starches.

Starch‐stärke, 54(10), 469-475.

165

Fujii, M., & Kawamura, Y. (1985). Synergistic action of α‐amylase and glucoamylase

on hydrolysis of starch. Biotechnology and Bioengineering, 27(3), 260-265.

Furuta, H., & Maeda, H. (1999). Rheological properties of water-soluble soybean

polysaccharides extracted under weak acidic condition. Food Hydrocolloids,

13(3), 267-274.

Garcıa-Ochoa, F., Santos, V., Casas, J., & Gomez, E. (2000). Xanthan gum:

Production, recovery, and properties. Biotechnology Advances, 18(7), 549-579.

Gidley, M. J. (2013). Hydrocolloids in the digestive tract and related health

implications. Current Opinion in Colloid & Interface Science, 18(4), 371-378.

Gonera, A., & Cornillon, P. (2002). Gelatinization of starch/gum/sugar systems

studied by using DSC, NMR, and CSLM. Starch‐stärke, 54(11), 508-516.

Goñi, I., Garcia-Alonso, A., & Saura-Calixto, F. (1997). A starch hydrolysis

procedure to estimate glycemic index. Nutrition Research, 17(3), 427-437.

Granfeldt, Y., & Björck, I. (1991). Glycemic response to starch in pasta: A study of

mechanisms of limited enzyme availability. Journal of Cereal Science, 14(1), 47-

61.

Gråsten, S., Liukkonen, K., Chrevatidis, A., El-Nezami, H., Poutanen, K., &

Mykkänen, H. (2003). Effects of wheat pentosan and inulin on the metabolic

activity of fecal microbiota and on bowel function in healthy humans. Nutrition

Research, 23(11), 1503-1514.

166

Gray, G. M. (1992). Starch digestion and absorption in nonruminants. The Journal of

Nutrition, 122(1), 172-177.

Guerra, A., Etienne-Mesmin, L., Livrelli, V., Denis, S., Blanquet-Diot, S., & Alric, M.

(2012). Relevance and challenges in modeling human gastric and small intestinal

digestion. Trends in Biotechnology, 30(11), 591-600.

Gularte, M. A., & Rosell, C. M. (2011). Physicochemical properties and enzymatic

hydrolysis of different starches in the presence of hydrocolloids. Carbohydrate

Polymers, 85(1), 237-244.

Gunness, P., Flanagan, B. M., Shelat, K., Gilbert, R. G., & Gidley, M. J. (2012).

Kinetic analysis of bile salt passage across a dialysis membrane in the presence of

cereal soluble dietary fibre polymers. Food Chemistry, 134(4), 2007-2013.

Gunness, P., & Gidley, M. J. (2010). Mechanisms underlying the cholesterol-lowering

properties of soluble dietary fibre polysaccharides. Food & Function, 1(2), 149-

155.

Guyton, A., & Hall, J. (2010). Textbook of medical physiology (12th ed.). Saunders,

London: Elsevier.

Hall, C., Tulbek, M. C., & Xu, Y. (2006). Flaxseed. Advances in Food and Nutrition

Research, 51, 1-97.

167

Halton, T. L., & Hu, F. B. (2004). The effects of high protein diets on thermogenesis,

satiety and weight loss: A critical review. Journal of the American College of

Nutrition, 23(5), 373-385.

Hasjim, J., Lavau, G. C., Gidley, M. J., & Gilbert, R. G. (2010). In vivo and in vitro

starch digestion: Are current in vitro techniques adequate? Biomacromolecules,

11(12), 3600-3608.

Health Canada and Statistics Canada. (2004). Canadian community health survey,

cycle 2.2, nutrition. provincial, regional and national summary data tables.

Retrieved September 11, 2014, from http://www.hc-sc.gc.ca/fn-

an/surveill/nutrition/commun/cchs_focus-volet_escc-eng.php

Health Canada. (2010). Proposed Policy: Definition and Energy Value for Dietary

Fibre. Retrieved September 11, 2014, from http://www.hc-sc.gc.ca/fn-

an/alt_formats/pdf/consult/policy-Definition-politique-eng.pdf

Health and Welfare Canada (1985). Report of the Expert Advisory Committee on

Dietary Fibre. Ottawa: Supply and Services Canada

Hengst, C., Ptok, S., Roessler, A., Fechner, A., & Jahreis, G. (2008). Effects of

polydextrose supplementation on different faecal parameters in healthy

volunteers. International Journal of Food Sciences and Nutrition, 60(s5), 96-105.

HIPSLEY, E. H. (1953). Dietary "fibre" and pregnancy toxaemia. British Medical

Journal, 2(4833), 420-422.

http://www.hc-sc.gc.ca/fn-an/surveill/nutrition/commun/cchs_focus-volet_escc-eng.php

http://www.hc-sc.gc.ca/fn-an/surveill/nutrition/commun/cchs_focus-volet_escc-eng.php

http://www.hc-sc.gc.ca/fn-an/alt_formats/pdf/consult/policy-Definition-politique-eng.pdf

http://www.hc-sc.gc.ca/fn-an/alt_formats/pdf/consult/policy-Definition-politique-eng.pdf

168

Hollebeeck, S., Borlon, F., Schneider, Y., Larondelle, Y., & Rogez, H. (2013).

Development of a standardised human in vitro digestion protocol based on

macronutrient digestion using response surface methodology. Food Chemistry,

138(2), 1936-1944.

Homann, H. H., Kemen, M., Fuessenich, C., Senkal, M., & Zumtobel, V. (1994).

Reduction in diarrhea incidence by soluble fiber in patients receiving total or

supplemental enteral nutrition. JPEN.Journal of Parenteral and Enteral

Nutrition, 18(6), 486-490.

Hur, S. J., Lim, B. O., Decker, E. A., & McClements, D. J. (2011). In vitro human

digestion models for food applications. Food Chemistry, 125(1), 1-12.

Institute of Medicine of the National Academies. (2009). Proposed definition of

dietary fibre. Retrieved September 10, 2014, from

http://www.iom.edu/Activities/Nutrition/DietaryFiberDef.aspx

J Wood, P., Braaten, J. T., Scott, F. W., Riedel, K. D., Wolynetz, M. S., & Collins, M.

W. (1994). Effect of dose and modification of viscous properties of oat gum on

plasma glucose and insulin following an oral glucose load. British Journal of

Nutrition, 72(05), 731-743.

Jarjis, H. A., Blackburn, N. A., Redfern, J. S., & Read, N. W. (1984). The effect of

ispaghula (fybogel and metamucil) and guar gum on glucose tolerance in man.

British Journal of Nutrition, 51(03), 371-378.

http://www.iom.edu/Activities/Nutrition/DietaryFiberDef.aspx

169

Jenkins, D. J., Kendall, C. W., Axelsen, M., Augustin, L. S., & Vuksan, V. (2000).

Viscous and nonviscous fibres, nonabsorbable and low glycaemic index

carbohydrates, blood lipids and coronary heart disease. Current Opinion in

Lipidology, 11(1), 49-56.

Jenkins, A. L., Jenkins, D. J., Zdravkovic, U., Wursch, P., & Vuksan, V. (2002).

Depression of the glycemic index by high levels of beta-glucan fiber in two

functional foods tested in type 2 diabetes. European Journal of Clinical

Nutrition, 56(7), 622-628.

Jenkins, D. J., Wolever, T. M., Taylor, R. H., Barker, H., Fielden, H., Baldwin, J. M.,

et al. (1981). Glycemic index of foods: A physiological basis for carbohydrate

exchange. The American Journal of Clinical Nutrition, 34(3), 362-366.

Jenkins, D. J. A., Leeds, A. R., Gassull, M. A., Cochet, B., & Alberti, K. G. M. M.

(1977). Decrease in postprandial insulin and glucose concentrations by guar and

pectin. Annals of Internal Medicine, 86(1), 20-23.

Jie, Z., Bang-Yao, L., Ming-Jie, X., Hai-Wei, L., Zu-Kang, Z., Ting-Song, W., et al.

(2000). Studies on the effects of polydextrose intake on physiologic functions in

chinese people. The American Journal of Clinical Nutrition, 72(6), 1503-1509.

Johnson, L. R. (1991). Gastrointestinal physiology. Missouri, St. Louis: Mosby Year

Book.

170

Johnson, L. R. (2014). Gastrointestinal physiology (8th ed.). Philadelphia, PA:

Elsevier Mosby.

Jones, J. R., Lineback, D. M., & Levine, M. J. (2006). Dietary reference intakes:

Implications for fiber labeling and consumption: A summary of the international

life sciences institute north america fiber workshop, june 1–2, 2004, washington,

DC. Nutrition Reviews, 64(1), 31-38.

Judd, P. A., & Ellis, P. R. (2005). 14 plant polysaccharides in the prevention and

treatment of diabetes mellitus. Traditional Medicines for Modern Times:

Antidiabetic Plants, 257.

Juvonen, K. R., Purhonen, A. K., Salmenkallio-Marttila, M., Lahteenmaki, L.,

Laaksonen, D. E., Herzig, K. H., et al. (2009). Viscosity of oat bran-enriched

beverages influences gastrointestinal hormonal responses in healthy humans. The

Journal of Nutrition, 139(3), 461-466.

Kahlon, T., Chiu, M., & Chapman, M. (2009). In vitro bile-acid-binding of whole vs.

pearled wheat grain. Cereal Chemistry, 86(3), 329-332.

Kang, K., & Pettit, D. (1993). Industrial gums. Polysaccharides and their Derivatives,

Kellett, G. L., & Brot-Laroche, E. (2005). Apical GLUT2: A major pathway of

intestinal sugar absorption. Diabetes, 54(10), 3056-3062.

Kendall, C. W., Esfahani, A., & Jenkins, D. J. (2010). The link between dietary fibre

and human health. Food Hydrocolloids, 24(1), 42-48.

171

Kerr, W., Ward, C., McWatters, K., & Resurreccion, A. A.(2000). effect of milling

and particle size on functionality and physicochemical properties of cowpea

flour. Cereal Chemistry, 77(2), 213-219.

Kim, C., & Yoo, B. (2006). Rheological properties of rice starch–xanthan gum

mixtures. Journal of Food Engineering, 75(1), 120-128.

Kristensen, M., & Jensen, M. G. (2011). Dietary fibres in the regulation of appetite

and food intake. importance of viscosity. Appetite, 56(1), 65-70.

Kuchel, P.W., & Ralston, G.B. (1997). Enzyme kinetics. In Schaum’s outline of

theory and problems in biochemistry (2nd

ed., pp. 251-274). New York, NY:

McGraw-Hill.

Kulicke, W., Eidam, D., Kath, F., Kix, M., & Kull, A. H. (1996). Hydrocolloids and

rheology: Regulation of Visco‐elastic characteristics of waxy rice starch in

mixtures with galactomannans. Starch‐stärke, 48(3), 105-114.

Kwong, M. G., Wolever, T. M., Brummer, Y., & Tosh, S. M. (2013). Attenuation of

glycemic responses by oat β-glucan solutions and viscoelastic gels is dependent

on molecular weight distribution. Food & Function, 4(3), 401-408.

Lee, R. E., Manthey, F. A., & Hall III, C. A. (2003). Effects of boiling, refrigerating,

and microwave heating on cooked quality and stability of lipids in macaroni

containing ground flaxseed. Cereal Chemistry, 80(5), 570-574.

172

Logan, K., Wright, A. J., & Goff, H. D. (2015). Correlating the structure and in vitro

digestion viscosities of different pectin fibers to in vivo human satiety. Food &

Function, 6(1), 62-70.

Mahasukhonthachat, K., Sopade, P., & Gidley, M. (2010). Kinetics of starch digestion

in sorghum as affected by particle size. Journal of Food Engineering, 96(1), 18-

28.

Maljaars, J., Peters, H., & Masclee, A. (2007). Review article: The gastrointestinal

tract: Neuroendocrine regulation of satiety and food intake. Alimentary

Pharmacology & Therapeutics, 26(s2), 241-250.

Mälkki, Y. (2001). Physical properties of dietary fiber as keys to physiological

functions. Cereal Foods World, 46(5), 196-199.

Mao, Y., & McClements, D. J. (2012). Influence of electrostatic heteroaggregation of

lipid droplets on their stability and digestibility under simulated gastrointestinal

conditions. Food & Function, 3(10), 1025-1034.

Marciani, L., Gowland, P. A., Fillery-Travis, A., Manoj, P., Wright, J., Smith, A., et

al. (2001). Assessment of antral grinding of a model solid meal with echo-planar

imaging. American Journal of Physiology.Gastrointestinal and Liver Physiology,

280(5), G844-9.

Marciani, L., Gowland, P. A., Spiller, R. C., Manoj, P., Moore, R. J., Young, P., et al.

(2000). Gastric response to increased meal viscosity assessed by echo-planar

173

magnetic resonance imaging in humans. The Journal of Nutrition, 130(1), 122-

127.

Marlett, J. A., & Fischer, M. H. (2003). The active fraction of psyllium seed husk.

Proceedings of the Nutrition Society, 62(01), 207-209.

Mazza, G., & Biliaderis, C. (1989). Functional properties of flax seed mucilage.

Journal of Food Science, 54(5), 1302-1305.

McDougall, G., Dobson, P., Shpiro, F., Smith, P., Stewart, D., & Fyffe, S. (2005,

August). Assessing bioavailability of soft fruit polyphenols in vitro. In I

International Symposium on Human Health Effects of Fruits and Vegetables 744

(pp. 135-148).

Milas, M., & Rinaudo, M. (1979). Conformational investigation on the bacterial

polysaccharide xanthan. Carbohydrate Research, 76(1), 189-196.

Miller, G. L. (1959). Use of dinitrosalicylic acid reagent for determination of reducing

sugar. Analytical Chemistry, 31(3), 426-428.

Minekus, M., Alminger, M., Alvito, P., Ballance, S., Bohn, T., Bourlieu, C., et al.

(2014). A standardised static in vitro digestion method suitable for food–an

international consensus. Food & Function, 5(6), 1113-1124.

Morris, D. H. (2001). Essential nutrients and other functional compounds in flaxseed.

Nutrition Today, 36(3), 159-162.

174

Nakamura, A., Fujii, N., Tobe, J., Adachi, N., & Hirotsuka, M. (2012).

Characterization and functional properties of soybean high-molecular-mass

polysaccharide complex. Food Hydrocolloids, 29(1), 75-84.

Nakamura, A., Furuta, H., Maeda, H., Takao, T., & Nagamatsu, Y. (2002). Structural

studies by stepwise enzymatic degradation of the main backbone of soybean

soluble polysaccharides consisting of galacturonan and rhamnogalacturonan.

Bioscience, Biotechnology, and Biochemistry, 66(6), 1301-1313.

Nakamura, A., Yoshida, R., Maeda, H., & Corredig, M. (2006). Soy soluble

polysaccharide stabilization at oil–water interfaces. Food Hydrocolloids, 20(2),

277-283.

Nalin, T., Koen, V., Weinstein, D. A., de Souza, C. F., Perry, I. D., van Wandelen, M.

T., et al. (2014). In vitro digestion of starches in a dynamic gastrointestinal

model: An innovative study to optimize dietary management of patients with

hepatic glycogen storage diseases. Journal of Inherited Metabolic Disease, , 1-8.

Nik, A. M., Wright, A. J., & Corredig, M. (2010). Surface adsorption alters the

susceptibility of whey proteins to pepsin-digestion. Journal of Colloid and

Interface Science, 344(2), 372-381.

Nilsson, A. C., Ostman, E. M., Holst, J. J., & Bjorck, I. M. (2008). Including

indigestible carbohydrates in the evening meal of healthy subjects improves

glucose tolerance, lowers inflammatory markers, and increases satiety after a

subsequent standardized breakfast. The Journal of Nutrition, 138(4), 732-739.

175

Oomen, A., Rompelberg, C., Bruil, M., Dobbe, C., Pereboom, D., & Sips, A. (2003).

Development of an in vitro digestion model for estimating the bioaccessibility of

soil contaminants. Archives of Environmental Contamination and Toxicology,

44(3), 0281-0287.

Owusu-Asiedu, A., Patience, J., Laarveld, B., Van Kessel, A., Simmins, P., & Zijlstra,

R. (2006). Effects of guar gum and cellulose on digesta passage rate, ileal

microbial populations, energy and protein digestibility, and performance of

grower pigs. Journal of Animal Science, 84(4), 843-852.

Panahi, S., Ezatagha, A., Temelli, F., Vasanthan, T., & Vuksan, V. (2007). β-Glucan

from two sources of oat concentrates affect postprandial glycemia in relation to

the level of viscosity. Journal of the American College of Nutrition, 26(6), 639-

644.

Pastrana, L., Gonzalez, M., Mirón, J., & Murado, M. (1998). A new device for

measuring diffusional restrictions and modelling substrate inhibition in a starch-

glucoamylase system. Biotechnology Letters, 20(2), 127-130.

Paxman, J., Richardson, J., Dettmar, P., & Corfe, B. (2008). Daily ingestion of

alginate reduces energy intake in free-living subjects. Appetite, 51(3), 713-719.

Phillips, G. O. (2013). Dietary fibre: A chemical category or a health ingredient?

Bioactive Carbohydrates and Dietary Fibre, 1(1), 3-9.

176

Public Health Agency of Canada. (2009). Tracking heart disease and stroke in

Canada. Retrieved September 10, 2014, from http://www.phac-

aspc.gc.ca/publicat/2009/cvd-avc/pdf/cvd-avs-2009-eng.pdf

Public Health Agency of Canada. (2013). Preventing chronic disease strategic plan

2013-2016. Retrieved September 10, 2014, from http://www.phac-aspc.gc.ca/cd-

mc/diabetes-diabete/strategy_plan-plan_strategique-eng.php

Qian, K., Cui, S., Wu, Y., & Goff, H. (2012). Flaxseed gum from flaxseed hulls:

Extraction, fractionation, and characterization. Food Hydrocolloids, 28(2), 275-

283.

Ramdath, D. D., Padhi, E., Hawke, A., Sivaramalingam, T., & Tsao, R. (2014). The

glycemic index of pigmented potatoes is related to their polyphenol content. Food

& function, 5(5), 909-915.

Redard, C. L., Davis, P. A., & Schneeman, B. O. (1990). Dietary fiber and gender:

Effect on postprandial lipemia. The American Journal of Clinical Nutrition,

52(5), 837-845.

Regand, A., Chowdhury, Z., Tosh, S. M., Wolever, T. M., & Wood, P. (2011). The

molecular weight, solubility and viscosity of oat beta-glucan affect human

glycemic response by modifying starch digestibility. Food Chemistry, 129(2),

297-304.

http://www.phac-aspc.gc.ca/publicat/2009/cvd-avc/pdf/cvd-avs-2009-eng.pdf

http://www.phac-aspc.gc.ca/publicat/2009/cvd-avc/pdf/cvd-avs-2009-eng.pdf

http://www.phac-aspc.gc.ca/cd-mc/diabetes-diabete/strategy_plan-plan_strategique-eng.php

http://www.phac-aspc.gc.ca/cd-mc/diabetes-diabete/strategy_plan-plan_strategique-eng.php

177

Renou, F., Petibon, O., Malhiac, C., & Grisel, M. (2013). Effect of xanthan structure

on its interaction with locust bean gum: Toward prediction of rheological

properties. Food Hydrocolloids, 32(2), 331-340.

Rinaudo, M., & Milas, M. (1982). Xanthan properties in aqueous solution.


Roberfroid, M. B., Van Loo, J. A., & Gibson, G. R. (1998). The bifidogenic nature of

chicory inulin and its hydrolysis products. The Journal of Nutrition, 128(1), 11-

19.

Roberts, F., Smith, H., Low, A., Ellis, P., Morris, E., & Sambrook, I. (1989).

Influence of guar gum flour of different molecular weights on viscosity of jejunal

digesta in the pig. Proc Nutr Soc, 49, 53A.

Roberts, K. (2011). The physiological and rheological effects of foods supplemented

with guar gum. Food Research International, 44(5), 1109-1114.

Sajjan, S. U., & Rao, M. R. (1987). Effect of hydrocolloids on the rheological

properties of wheat starch. Carbohydrate Polymers, 7(5), 395-402.

Sasaki, T., & Kohyama, K. (2011). Effect of non-starch polysaccharides on the in

vitro digestibility and rheological properties of rice starch gel. Food Chemistry,

127(2), 541-546.

178

Sasaki, T., & Kohyama, K. (2012). Influence of non-starch polysaccharides on the in

vitro digestibility and viscosity of starch suspensions. Food Chemistry, 133(4),

1420-1426.

Schneeman, B. O. (1998). Dietary fiber and gastrointestinal function. Nutrition

Research, 18(4), 625-632.

Schulze, M. B., & Hu, F. B. (2005). Primary prevention of diabetes: What can be done

and how much can be prevented? Annu.Rev.Public Health, 26, 445-467.

Shimotoyodome, A., Meguro, S., Hase, T., Tokimitsu, I., & Sakata, T. (2001).

Sulfated polysaccharides, but not cellulose, increase colonic mucus in rats with

loperamide-induced constipation. Digestive Diseases and Sciences, 46(7), 1482-

1489.

Singh, J., Dartois, A., & Kaur, L. (2010). Starch digestibility in food matrix: A

review. Trends in Food Science & Technology, 21(4), 168-180.

Slavin, J., & Green, H. (2007). Dietary fibre and satiety. Nutrition Bulletin, 32(s1),

32-42.

Slavin, J. (2013). Fiber and prebiotics: Mechanisms and health benefits. Nutrients,

5(4), 1417-1435.

Slavin, J., & Jacobs Jr, D. R. (2010). Dietary fiber: All fibers are not alike. Nutrition

guide for physicians (pp. 13-24) Springer.

179

Southgate, D. (1969). Determination of carbohydrates in foods II.—Unavailable

carbohydrates. Journal of the Science of Food and Agriculture, 20(6), 331-335.

Sreerama, Y. N., Sashikala, V. B., & Pratape, V. M. (2012). Phenolic compounds in

cowpea and horse gram flours in comparison to chickpea flour: Evaluation of

their antioxidant and enzyme inhibitory properties associated with hyperglycemia

and hypertension. Food Chemistry, 133(1), 156-162.

Steffe, J. F. (1996). Rheological methods in food process engineering Freeman press.

Streppel, M. T., Arends, L. R., van’t Veer, P., Grobbee, D. E., & Geleijnse, J. M.

(2005). Dietary fiber and blood pressure: A meta-analysis of randomized

placebo-controlled trials. Archives of Internal Medicine, 165(2), 150-156.

Sudhakar, V., Singhal, R., & Kulkarni, P. (1996). Starch-galactomannan interactions:

Functionality and rheological aspects. Food Chemistry, 55(3), 259-264.

Symons, L., & Brennan, C. (2004). The influence of (1→ 3)(1→ 4)‐β‐D‐Glucan‐rich

fractions from barley on the physicochemical properties and in vitro reducing

sugar release of white wheat breads. Journal of Food Science, 69(6), C463-C467.

Takahashi, H., Wako, N., Okubo, T., Ishihara, N., Yamanaka, J., & Yamamoto, T.

(1994). Influence of partially hydrolyzed guar gum on constipation in women.

Journal of Nutritional Science and Vitaminology, 40(3), 251-259.

Takeda, Y., Hizukuri, S., & Juliano, B. O. (1986). Purification and structure of

amylose from rice starch. Carbohydrate Research, 148(2), 299-308.

180

Teaford, M. F., & Ungar, P. S. (2000). Diet and the evolution of the earliest human

ancestors. Proceedings of the National Academy of Sciences of the United States

of America, 97(25), 13506-13511.

Thakur, G., Mitra, A., Pal, K., & Rousseau, D. (2009). Effect of flaxseed gum on

reduction of blood glucose and cholesterol in type 2 diabetic patients.

International Journal of Food Sciences and Nutrition, 60(s6), 126-136.

Tharakan, A., Norton, I., Fryer, P., & Bakalis, S. (2010). Mass transfer and nutrient

absorption in a simulated model of small intestine. Journal of Food Science,

75(6), E339-E346.

Tharakan, A., Norton, I., Fryer, P., & Bakalis, S. (2010). Mass transfer and nutrient

absorption in a simulated model of small intestine. Journal of Food Science,

75(6), E339-E346.

Threapleton, D. E., Greenwood, D. C., Evans, C. E., Cleghorn, C. L., Nykjaer, C.,

Woodhead, C., et al. (2013). Dietary fibre intake and risk of cardiovascular

disease: Systematic review and meta-analysis. BMJ (Clinical Research Ed.), 347,

f6879.

Topping, D. L., & Clifton, P. M. (2001). Short-chain fatty acids and human colonic

function: Roles of resistant starch and nonstarch polysaccharides. Physiological

Reviews, 81(3), 1031-1064.

181

Tsai, A. C., Mott, E. L., Owen, G. M., Bennick, M. R., Lo, G. S., & Steinke, F. H.

(1983). Effects of soy polysaccharide on gastrointestinal functions, nutrient

balance, steroid excretions, glucose tolerance, serum lipids, and other parameters

in humans. The American Journal of Clinical Nutrition, 38(4), 504-511.

Tungland, B., & Meyer, D. (2002). Nondigestible oligo‐and polysaccharides (dietary

fiber): Their physiology and role in human health and food. Comprehensive

Reviews in Food Science and Food Safety, 1(3), 90-109.

U.S. Food and Drug Administration (FDA). (2013). Guidance for industry: A food

labeling guide (11. appendix C: Health claims). Retrieved September 25, 2014,

from

http://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryIn

formation/LabelingNutrition/ucm064919.htm

Urlacher, B., & Noble, O. (1997). Xanthan gum. Thickening and gelling agents for

food (pp. 284-311) Springer.

Varatharajan, V., Hoover, R., Li, J., Vasanthan, T., Nantanga, K., Seetharaman, K., et

al. (2011). Impact of structural changes due to heat-moisture treatment at

different temperatures on the susceptibility of normal and waxy potato starches

towards hydrolysis by porcine pancreatic alpha amylase. Food Research

International, 44(9), 2594-2606.

Vardakou, M., Mercuri, A., Barker, S. A., Craig, D. Q., Faulks, R. M., & Wickham,

M. S. (2011). Achieving antral grinding forces in biorelevant in vitro models:

http://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/LabelingNutrition/ucm064919.htm

http://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/LabelingNutrition/ucm064919.htm

182

Comparing the USP dissolution apparatus II and the dynamic gastric model with

human in vivo data. Aaps Pharmscitech, 12(2), 620-626.

Vieira, F. C., & Sarmento, S. (2008). Heat‐Moisture treatment and enzymatic

digestibility of peruvian carrot, sweet potato and ginger starches. Starch‐stärke,

60(5), 223-232.

Wang, J., Zeng, A., Liu, Z., & Yuan, X. (2006). Kinetics of glucoamylase hydrolysis

of corn starch. Journal of Chemical Technology and Biotechnology, 81(4), 727-

729.

White, G., Doucet, F., Hill, S., & Wiseman, J. (2008). Physicochemical changes to

starch granules during micronisation and extrusion processing of wheat, and their

implications for starch digestibility in the newly weaned piglet. Animal, 2:9,

1312–1323

Wickham, M., Faulks, R., Mann, J., & Mandalari, G. (2012). The design, operation,

and application of a dynamic gastric model. Dissolution Technologies, 19(3), 15-

22.

Wielinga W. (2010). Seed Gums (From Imeson A’s Food Stabilisers, Thickeners and

Gelling Agents). Ames, Iowa: Wiley-Blackwell Publishing Ltd.

Wielinga W. & Maehall AG. (2000). Galactomannans (From Phillips’ & Williams’

Handbook of Hydrocolloids). NY, New York: CRC Press.

183

Wiseman, J. (2006). Variations in starch digestibility in non-ruminants. Animal Feed

Science and Technology, 130(1), 66-77.

Wolever, T. M., Tosh, S. M., Gibbs, A. L., Brand-Miller, J., Duncan, A. M., Hart, V.,

et al. (2010). Physicochemical properties of oat beta-glucan influence its ability

to reduce serum LDL cholesterol in humans: A randomized clinical trial. The

American Journal of Clinical Nutrition, 92(4), 723-732.

Wood, P., Braaten, J. T., Scott, F.W., Riedel, K.D., Wolynetz, M.S., & Collins, M.W.

(1994). Effect of dose and modification of viscous properties of oat gum on

plasma glucose and insulin following an oral glucose load. British Journal of

Nutrition, 72:731-743.

Wood, P., Beer, M., & Butler, G. (2000). Evaluation of role of concentration and

molecular weight of oat β-glucan in determining effect of viscosity on plasma

glucose and insulin following an oral glucose load. British Journal of Nutrition,

84(01), 19-23.

Woolnough, J. W., Monro, J. A., Brennan, C. S., & Bird, A. R. (2008). Simulating

human carbohydrate digestion in vitro: A review of methods and the need for

standardisation. International Journal of Food Science & Technology, 43(12),

2245-2256.

World Health Organization. (2003). Screening for type 2 diabetes. Retrieved

September 10 2014, from

http://www.who.int/diabetes/publications/en/screening_mnc03.pdf

http://www.who.int/diabetes/publications/en/screening_mnc03.pdf

184

Wrick, K. L., Robertson, J. B., Van Soest, P. J., Lewis, B. A., Rivers, J. M., Roe, D.

A., et al. (1983). The influence of dietary fiber source on human intestinal transit

and stool output. The Journal of Nutrition, 113(8), 1464-1479.

Wursch, P., & Pi-Sunyer, F. X. (1997). The role of viscous soluble fiber in the

metabolic control of diabetes. A review with special emphasis on cereals rich in

beta-glucan. Diabetes Care, 20(11), 1774-1780.

Yamatoya, K., Kuwano, K., & Suzuki, J. (1997). Effects of hydrolyzed guar gum on

cholesterol and glucose in humans. Food Hydrocolloids, 11(2), 239-242.

Zhang, J., Cui, J. H., Yin, T., Sun, L., & Li, G. (2013). Activated effect of lignin on α-

amylase. Food Chemistry, 141(3), 2229-2237.